Words and The Company They Keep (homework 2)¶

Solution to the second homework for the NPFL067 course.

In [1]:
source_files = [
    "CZ.txt",
    "EN.txt"
]
dir = "plain"

## load the text files
def load_text(filename, dir) -> str:
    with open(f"../texts/{dir}/{filename}", "r", encoding="utf-8") as f:
        return f.read()

texts = {file : load_text(file, dir).split('\n') for file in source_files}

for short_text, words in texts.items():
    print(f"Text: {short_text}, word-count: {len(words)}, unique-words: {len(set(words))}")

lexicons = {file : set(texts[file]) for file in texts}
alphabets = {file : set("".join(texts[file])) for file in texts}
Text: CZ.txt, word-count: 222413, unique-words: 42827
Text: EN.txt, word-count: 221099, unique-words: 9608
In [2]:
from typing import List, Dict, Set, Tuple, Any
from collections import Counter, defaultdict
import pandas as pd
from math import log2

Best Friends¶

Compute the pointwise mutual information for all the possible word pairs appearing consecutively in the data, disregarding pairs in which one or both words appear less than 10 times in the corpus, and sort the results from the best to the worst (did you get any negative values? Why?) Tabulate the results, and show the best 20 pairs for both data sets.

Do the same now but for distant words, i.e. words which are at least 1 word apart, but not farther than 50 words (both directions). Again, tabulate the results, and show the best 20 pairs for both data sets.

In [3]:
## obtain the count for all word pairs appering consecutively
def count_word_pairs(text : List[str]) -> Counter:
    return Counter(zip(text, text[1:]))

def count_word_surrounding_pairs(text : List[str], min_distance : int = 2, max_distance : int = 50) -> Counter:
    counter = Counter()
    ## (both derections)
    for dist in range(min_distance, max_distance + 1):
        counter.update(zip(text, text[dist:]))
        counter.update(zip(text[dist:], text))
    return counter

## cut out less frequent word pairs (less than 10)
def cut_non_frequent_words(text : List['str']) -> List['str']:
    limit = 10
    word_count = Counter(text)
    return [word for word in text if word_count[word] >= limit]

## show top 20 most frequent word pairs
def show_top_word_pairs(counter : Counter, n : int = 20) -> None:
    for pair, count in counter.most_common(n):
        print(f"\t'{pair}' - #{count}")
In [4]:
tables = []
for short_text in texts:
    for count_func, shortcut in [
        (count_word_pairs, "Consecutive"),
        (count_word_surrounding_pairs, "Surrounding")
    ]:
        source_text = cut_non_frequent_words(texts[short_text])
        word_pairs : Counter = count_func(source_text)
        df = pd.DataFrame(word_pairs.items(), columns=[f"Pair", f"Count"]).nlargest(20, f"Count").reset_index(drop=True)
        tables.append(pd.DataFrame([""]*20, columns=[f"({shortcut}) {short_text}"]))
        tables.append(df)

print("Top 20 most frequent word pairs - counts")
display(pd.concat(tables, axis=1))
Top 20 most frequent word pairs - counts
(Consecutive) CZ.txt Pair Count (Surrounding) CZ.txt Pair Count (Consecutive) EN.txt Pair Count (Surrounding) EN.txt Pair Count
0 (,, že) 1615 (,, ,) 135650 (of, the) 2770 (,, ,) 106898
1 (., ,) 834 (., .) 118008 (,, and) 2004 (the, ,) 95306
2 (., .) 774 (., ,) 117613 (in, the) 1387 (,, the) 95306
3 (,, ,) 727 (,, .) 117613 (the, same) 938 (the, the) 91746
4 (., ") 631 (:, :) 42618 (to, the) 891 (,, of) 67540
5 (,, a) 520 (:, ,) 39492 (;, and) 769 (of, ,) 67540
6 (,, ale) 452 (,, :) 39492 (,, as) 767 (the, of) 64310
7 ((, )) 417 (a, ,) 39208 (,, the) 714 (of, the) 64310
8 (:, 1) 412 (,, a) 39208 (have, been) 630 (of, of) 45186
9 (), ,) 411 (a, .) 37133 (the, of) 629 (., ,) 41251
10 (., V) 403 (., a) 37133 (on, the) 625 (,, .) 41251
11 (a, .) 396 (v, ,) 34673 (that, the) 567 (and, ,) 38733
12 (., se) 392 (,, v) 34673 (., The) 528 (,, and) 38733
13 (,, který) 382 (v, .) 34066 (,, which) 527 (., the) 37235
14 (!, !) 379 (., v) 34066 (,, or) 464 (the, .) 37235
15 (,, které) 376 (., :) 32939 (,, that) 462 (and, the) 36513
16 (:, 0) 342 (:, .) 32939 (,, in) 461 (the, and) 36513
17 (., :) 329 (,, se) 30789 (., ,) 419 (,, in) 33828
18 (1, :) 327 (se, ,) 30789 (and, the) 398 (in, ,) 33828
19 (", ") 320 (., se) 28024 (species, ,) 396 (,, to) 32485
In [5]:
## compute the pointwise mutual information for the word pairs
results : Dict[str, Dict[str, Dict[Any, float]]] = {}
for short_text in texts:
    results[short_text] = {}
    for count_func, label in [
        (count_word_pairs, "Consecutive"),
        (count_word_surrounding_pairs, "Surrounding")
        ]:

        source_text = cut_non_frequent_words(texts[short_text])
        text_len = len(source_text)
        word_prob = {word : count / text_len for word, count in Counter(source_text).items()}

        pairs_prob = defaultdict(float)
        word_pairs : Counter = count_func(source_text)
        total_pairs = sum(word_pairs.values())
        for pair, count in word_pairs.items():
            pairs_prob[pair] = count / total_pairs

        results[short_text][label] = {pair : log2(pairs_prob[pair] / (word_prob[pair[0]] * word_prob[pair[1]])) for pair in word_pairs}

## show the top n most significant word pairs
n = 20
tables_top = []
tables_worst = []
for short_text in results:
    for label in results[short_text]:
        df_top = pd.DataFrame(results[short_text][label].items(), columns=[f"Pair", f"PMI"]).nlargest(n, f"PMI").reset_index(drop=True)
        tables_top.append(pd.DataFrame([""]*n, columns=[f"({label}) {short_text}"]))
        tables_top.append(df_top)
        df_worst = pd.DataFrame(results[short_text][label].items(), columns=[f"Pair", f"PMI"]).nsmallest(n, f"PMI").reset_index(drop=True)
        tables_worst.append(pd.DataFrame([""]*n, columns=[f"({label}) {short_text}"]))
        tables_worst.append(df_worst)


print(f"Top {n} most significant word pairs - PMI")
display(pd.concat(tables_top, axis=1))

print(f"Worst {n} most significant word pairs - PMI")
display(pd.concat(tables_worst, axis=1))
Top 20 most significant word pairs - PMI
(Consecutive) CZ.txt Pair PMI (Surrounding) CZ.txt Pair PMI (Consecutive) EN.txt Pair PMI (Surrounding) EN.txt Pair PMI
0 (Hamburger, SV) 13.712115 (výher, výher) 9.571579 (La, Plata) 14.042450 (dried, floated) 8.690953
1 (Los, Angeles) 13.485607 (žel, žel) 8.974643 (Asa, Gray) 13.904947 (floated, dried) 8.690953
2 (Johna, Newcomba) 13.186047 (Sandžaku, Sandžaku) 8.834613 (Fritz, Muller) 13.235095 (germinated, dried) 8.318984
3 (Č., Budějovice) 13.056763 (Petrof, Petrof) 8.723582 (worth, while) 13.205949 (dried, germinated) 8.318984
4 (série, ATP) 12.892133 (CIA, CIA) 8.628162 (faced, tumbler) 13.135560 (floated, floated) 8.316887
5 (turnajové, série) 12.857576 (IFS, IFS) 8.401654 (lowly, organised) 13.089979 (dried, dried) 8.176380
6 (Tomáš, Ježek) 12.852146 (13h, 13h) 8.401654 (Malay, Archipelago) 12.983557 (germinated, floated) 8.083964
7 (Lidové, noviny) 12.753087 (Bělehrad, Benfica) 8.375181 (shoulder, stripe) 12.926973 (floated, germinated) 8.083964
8 (Lidových, novin) 12.694193 (Benfica, Bělehrad) 8.375181 (Great, Britain) 12.787636 (clover, clover) 8.006455
9 (veřejného, mínění) 12.485607 (silniční, silniční) 8.249650 (THE, THE) 12.779416 (heath, heath) 7.933930
10 (teplota, minus) 12.404687 (Divadlo, Divadlo) 8.170699 (United, States) 12.720522 (days, dried) 7.927992
11 (Ján, Čarnogurský) 12.378692 (IV, výher) 8.086152 (deserves, notice) 12.605045 (dried, days) 7.927992
12 (jaderné, zbraně) 12.378692 (výher, IV) 8.086152 (CHAPTER, THE) 12.457488 (floated, days) 7.893582
13 (Milan, Máčala) 12.320976 (km, žel) 8.064841 (English, carrier) 12.398594 (days, floated) 7.893582
14 (lidských, práv) 12.286042 (žel, km) 8.064841 (specially, endowed) 12.274896 (vibracula, vibracula) 7.842956
15 (společném, státě) 12.131599 (13h, zataženo) 7.962163 (Sir, J) 12.250443 (Pacific, eastern) 7.759900
16 (akciových, společností) 12.115657 (zataženo, 13h) 7.962163 (branched, off) 12.250443 (eastern, Pacific) 7.759900
17 (Pohár, UEFA) 12.048543 (OKD, OKD) 7.927722 (de, Candolle) 12.235095 (masters, slaves) 7.758308
18 (privatizačních, projektů) 12.038842 (13h, st) 7.886620 (mental, qualities) 12.235095 (slaves, masters) 7.758308
19 (George, Bushe) 12.026175 (st, 13h) 7.886620 (Galapagos, Archipelago) 12.218022 (stripe, shoulder) 7.753014
Worst 20 most significant word pairs - PMI
(Consecutive) CZ.txt Pair PMI (Surrounding) CZ.txt Pair PMI (Consecutive) EN.txt Pair PMI (Surrounding) EN.txt Pair PMI
0 (., že) -6.200420 (jednání, 3) -6.274230 (the, a) -7.684486 (forms, instincts) -4.335294
1 (:, že) -5.287548 (3, jednání) -6.274230 (a, the) -7.684486 (instincts, forms) -4.335294
2 (že, :) -5.287548 (že, =) -6.243805 (in, in) -6.806685 (colour, will) -4.308060
3 (:, s) -4.742022 (=, že) -6.243805 (the, the) -6.311216 (will, colour) -4.308060
4 (a, 1) -4.728166 (nebo, 0) -6.109001 (the, it) -6.296825 (fish, all) -4.235751
5 (,, Na) -4.609473 (0, nebo) -6.109001 (the, this) -6.223902 (all, fish) -4.235751
6 (Na, .) -4.516894 (2, strany) -5.888557 (same, the) -6.053185 (forms, food) -4.221338
7 (:, i) -4.503608 (strany, 2) -5.888557 (their, the) -5.949668 (food, forms) -4.221338
8 (!, .) -4.478124 (3, vlády) -5.842347 (., be) -5.852153 (wax, species) -4.176287
9 (., co) -4.432502 (vlády, 3) -5.842347 (., which) -5.618343 (species, wax) -4.176287
10 (a, 2) -4.360909 (této, 0) -5.703182 (of, we) -5.613237 (Europe, selection) -4.163019
11 ((, () -4.356932 (0, této) -5.703182 (;, to) -5.594396 (selection, Europe) -4.163019
12 (se, 1) -4.318905 (3, státu) -5.623178 (of, be) -5.582921 (comb, species) -4.105898
13 ((, že) -4.313363 (státu, 3) -5.623178 (., by) -5.569208 (species, comb) -4.105898
14 (v, by) -4.206675 (2, korun) -5.585630 (we, ,) -5.265299 (forms, side) -4.097602
15 (a, V) -4.197833 (korun, 2) -5.585630 (many, the) -5.234009 (side, forms) -4.097602
16 (:, k) -4.129483 (kteří, 4) -5.572841 (has, of) -5.226108 (flying, species) -4.031897
17 (kteří, .) -4.109124 (4, kteří) -5.572841 (this, the) -5.223902 (species, flying) -4.031897
18 (se, je) -4.044692 (4, u) -5.558268 (the, any) -5.218822 (early, no) -4.027751
19 (., roku) -4.011659 (u, 4) -5.558268 (at, of) -5.208164 (no, early) -4.027751

Intepretation¶

We can see from the obtained results that we got both negative and positive values. That illustrates well the following:

"Good collocation pairs have high PMI because the probability of co-occurrence is only slightly lower than the probabilities of occurrence of each word. Conversely, a pair of words whose probabilities of occurrence are considerably higher than their probability of co-occurrence gets a small PMI score."

Especially, pairs like "Lidové noviny" in Czech or "United States" in English represent very strong collocations used often with the second word (that is why their PMI is very high), while pairs like "na, se" in Czech or "to, and" in English are all but collocations (and that is why their PMI is very low).

We can also see PMI scores closer to 0 for surrounding pairs (compared to the neighbouring pairs) in the TOP chart. This might be because of the fact that the farther away the second word is, the less dependent (relatively to the first word in the text) the occurence of the second word is. And we know that the PMI is equal 0 if both X and Y are independant.

Word Classes¶

Compute a full class hierarchy of words using the first 8,000 words of those data, and only for words occurring 10 times or more (use the same setting for both languages). Ignore the other words for building the classes, but keep them in the data for the bigram counts and all the formulas that use them (including the Mutual Information, the interim sums in the "Tricks", etc.). For details on the algorithm, use the Brown et al. paper available form SIS; some formulas are wrong in the paper, however, so please see the corrections in the slides (formulas for Trick #4). Note the history of the merges, and attach it to your homework. Now run the same algorithm again, but stop when reaching 15 classes. Print out all the members of your 15 classes and attach them too.

Text preparation¶

In [8]:
from typing import List, Dict, Tuple, Set
from collections import Counter
import numpy as np
import math
import itertools
import pandas as pd

separator = '/'

## load the text files
def load_text(filename, dir) -> str:
    with open(f"../texts/{dir}/{filename}", "r", encoding="utf-8") as f:
        return f.read()

def load_ptg_text(text : str) -> List[str]:
    return text.split('\n')

def load_word_and_pos(ptg_text : List[str]) -> List[Tuple[str, str]]:
    return [(word.split(separator)[0], "".join(word.split(separator)[1:]))
            for word in ptg_text]

def separate_words_and_pos(ptg_text : List[Tuple[str, str]]) -> Tuple[List[str], List[str]]:
    return [word for word, _ in ptg_text], [pos for _, pos in ptg_text]

def get_unique_words(text : List[str]) -> Set[str]:
    return set(text)

Implementing merging classes with tricks from lecture¶

In [9]:
## TRICK 1 - bigram counts

def get_word_counts(text : List[str]) -> Counter:
    return Counter(text)

def get_bigram_counts(text : List[str]) -> Counter:
    return Counter(zip(text, text[1:]))

def get_left_unigram_counts(bigram_counter : Counter) -> Counter:
    unigram_counter = Counter()
    for (word1, word2), count in bigram_counter.items():
        unigram_counter[word1] += count
    return unigram_counter

def get_right_unigram_counts(bigram_counter : Counter) -> Counter:
    unigram_counter = Counter()
    for (word1, word2), count in bigram_counter.items():
        unigram_counter[word2] += count
    return unigram_counter

def total_number_of_words(text : List[str]) -> int:
    return len(text)

def total_number_of_unique_words(text : List[str]) -> int:
    return len(get_unique_words(text))

def arange_classes(unique_words : Set[str], FREQ_THRESH : int = 1, counter : Counter = None) -> np.ndarray:
    if (FREQ_THRESH > 1 and counter is not None):
        ## only words with frequency higher than K are considered
        return np.array([i for i, word in enumerate(unique_words) if counter[word] >= FREQ_THRESH], dtype=np.int64)
    else:
        ## all words are considered
        return np.arange(len(unique_words), dtype=np.int64)

def create_bigram_matrix(unique_words : List[str], bigram_counter : Counter, word_to_class_mapping : Dict[str, int]) -> np.ndarray:
    C = len(unique_words)
    bigram_matrix = np.zeros((C, C), dtype=np.int64)
    for (word1, word2), count in bigram_counter.items():
        bigram_matrix[word_to_class_mapping[word1], word_to_class_mapping[word2]] = count
    return bigram_matrix

def create_unigram_matrix(unique_words : List[str], unigram_counter : Counter, word_to_class_mapping : Dict[str, int]) -> np.ndarray:
    C = len(unique_words)
    unigram_matrix = np.zeros(C, dtype=np.int64)
    for word, count in unigram_counter.items():
        unigram_matrix[word_to_class_mapping[word]] = count
    return unigram_matrix

def create_word_to_class_mapping(unique_words : List[str]) -> Dict[str, int]:
    return {word : i for i, word in enumerate(unique_words)}

def create_class_to_word_mapping(unique_words : List[str]) -> Dict[int, List[str]]:
    return {i : [word] for i, word in enumerate(unique_words)}
In [10]:
## TRICK 2 - precompute q and s

def q(c_bi, c_l, c_r, N):
    if (c_l == 0 or c_r == 0 or c_bi == 0):
        return 0
    return (c_bi/N)*np.log2((N*c_bi)/(c_l*c_r))

def precompute_q_and_s(classes, c_bi, c_l, c_r, C, N, verbose : bool = False):
    if verbose:
        print(f"Computing 'Q'...")

    Q = np.zeros((C,C), dtype=float)
    for l in range(C):
        for r in range(C):
            Q[l, r] = q(c_bi[l, r], c_l[l], c_r[r], N)

    if verbose:
        print(f"Computing 'S'...")
        
    S = np.zeros((C), dtype=float)
    for cls in classes:
        S[cls] = (np.sum(Q[:, cls]) + np.sum(Q[cls,:]) - Q[cls, cls])

    return Q, S
In [11]:
## TRICK 3: sub and add
def sub(S, Q, a : int, b : int):
    return (S[a] + S[b] - Q[a, b] - Q[b, a])

def add(counts_current : tuple["np.ndarray", "np.ndarray", "np.ndarray"], N, C, a : int, b : int):
    """'c' is the class into which was 'a' and 'b' merged. The rule is to merge 'b' into 'a'."""
    c_bi, c_l, c_r = counts_current
    
    sums = np.sum([
        q(c_bi[x, a] + c_bi[x, b], c_l[x], c_r[a] + c_r[b], N) ## q(l, a+b)
        + q(c_bi[a, x] + c_bi[b, x], c_l[a] + c_l[b], c_r[x], N) ## q(a+b, r)
        for x in range(C) if (x != a and x != b)
    ])

    merged = q(c_bi[a, a] + c_bi[a, b] + c_bi[b, a] + c_bi[b, b], c_l[a] + c_l[b], c_r[a] + c_r[b], N) ## q(a+b, a+b)

    return (sums + merged)
In [12]:
## TRICK 4 - table of losses
def init_l(Q, S, N, C, counts_current : tuple["np.ndarray", "np.ndarray", "np.ndarray"], classes, verbose : bool = False):
    if verbose:
        print(f"Init of 'l'...")
        
    ## create table of losses for each possible pair (upper right triangle)
    L = np.zeros((C,C), dtype=float)
    for cls_one, cls_two in itertools.combinations(classes, 2):
        ## calculate the loss of merging these two classes
        L[cls_one, cls_two] = sub(S, Q, cls_one, cls_two) - add(counts_current, N, C, cls_one, cls_two)
        L[cls_two, cls_one] = L[cls_one, cls_two] ## because of symmetry
    return L

def find_best_pair(L, classes):
    minimal_loss = math.inf
    best_pair = (-1, -1)
    for i, j in itertools.combinations(classes, 2):
        if L[i, j] < minimal_loss:
            minimal_loss = L[i, j]
            best_pair = (i, j)
    return best_pair, minimal_loss

def mergeCounts(current_counts : tuple["np.ndarray", "np.ndarray", "np.ndarray"], a : int, b : int) -> tuple["np.ndarray", "np.ndarray", "np.ndarray"]:
    """Merges the class 'b' into the class 'a'. Adjusts the counts."""
    c_bi, c_l, c_r = current_counts
    c_bi_f = c_bi.copy()
    c_l_f = c_l.copy()
    c_r_f = c_r.copy()

    ## merge the counts
    c_bi_f[a, :] += c_bi_f[b, :]
    c_bi_f[:, a] += c_bi_f[:, b]
    c_l_f[a] += c_l_f[b]
    c_r_f[a] += c_r_f[b]

    ## zero the merged class
    c_bi_f[b, :] = np.zeros(c_bi_f.shape[0])
    c_bi_f[:, b] = np.zeros(c_bi_f.shape[0])
    c_l_f[b] = 0
    c_r_f[b] = 0

    return c_bi_f, c_l_f, c_r_f

def update(L, S, counts_current : tuple["np.ndarray", "np.ndarray", "np.ndarray"], counts_future : tuple["np.ndarray", "np.ndarray", "np.ndarray"], classes, N, a : int, b : int):
    c_bi, c_l, c_r = counts_current
    c_bi_f, c_l_f, c_r_f = counts_future

    S_f = S.copy()
    for i in classes:
        if i == b:
            continue
        S_f[i] = (S[i]
                  - q(c_bi[i, a], c_l[i], c_r[a], N) ## q(i, a)
                  - q(c_bi[a, i], c_l[a], c_r[i], N) ## q(a, i)
                  - q(c_bi[i, b], c_l[i], c_r[b], N) ## q(i, b)
                  - q(c_bi[b, i], c_l[b], c_r[i], N) ## q(b, i)
                  + q(c_bi_f[a, i], c_l_f[a], c_r_f[i], N) ## q(a, i)
                  + q(c_bi_f[i, a], c_l_f[i], c_r_f[a], N)) ## q(i, a)

    ## treated separately - like init
    S_f[a] = (np.sum([q(c_bi_f[x, a], c_l_f[x], c_r_f[a], N) for x in range(c_bi.shape[0])]) + np.sum([q(c_bi_f[a, x], c_l_f[a], c_r_f[x], N) for x in range(c_bi.shape[0])]) - q(c_bi_f[a, a], c_l_f[a], c_r_f[a], N))

    L_f = L.copy()
    
    for i, j in itertools.combinations(classes, 2):
        if (i == a and j not in [a, b]):
            ## treated separately - like init
            L_f[i, j] = (S_f[i]
                         + S_f[j]
                         - q(c_bi_f[i, j], c_l_f[i], c_r_f[j], N)
                         - q(c_bi_f[j, i], c_l_f[j], c_r_f[i], N)
                         - q(c_bi_f[i, i] + c_bi_f[i, j] + c_bi_f[j, i] + c_bi_f[j, j], c_l_f[i] + c_l_f[j], c_r_f[i] + c_r_f[j], N)
                         - np.sum([q(c_bi_f[x, i] + c_bi_f[x, j], c_l_f[x], c_r_f[i] + c_r_f[j], N) for x in range(c_bi.shape[0]) if x not in [i, j]])
                         - np.sum([q(c_bi_f[i, x] + c_bi_f[j, x], c_l_f[i] + c_l_f[j], c_r_f[x], N) for x in range(c_bi.shape[0]) if x not in [i, j]]))
        else:
            L_f[i, j] = (L[i, j] 
                        - S[i] 
                        + S_f[i] 
                        - S[j] 
                        + S_f[j]
                        + q(c_bi[i, a]+c_bi[j, a], c_l[i]+c_l[j], c_r[a], N)
                        + q(c_bi[a, i]+c_bi[a, j], c_l[a], c_r[i]+c_r[j], N)
                        + q(c_bi[i, b]+c_bi[j, b], c_l[i]+c_l[j], c_r[b], N)
                        + q(c_bi[b, i]+c_bi[b, j], c_l[b], c_r[i]+c_r[j], N)
                        - q(c_bi_f[i, a]+c_bi_f[j, a], c_l_f[i]+c_l_f[j], c_r_f[a], N)
                        - q(c_bi_f[a, i]+c_bi_f[a, j], c_l_f[a], c_r_f[i]+c_r_f[j], N))
        
        L_f[j, i] = L_f[i, j] ## because of symmetry

    return L_f, S_f
In [13]:
source_files = {
    "en" : "EN.txt",
    "cz" : "CZ.txt"
}

dir = "ptg"

def execute_merging(
        lang : str, 
        mode : str, 
        LIMIT : int, 
        MIN_FREQ : int,
        K : int,
        verbose : bool = False
        ) -> List[List[str]]:
    '''Returns a list of classes, each class is a list of words that were merged together.'''

    if verbose:
        print(f"Merging language: {lang}, mode: {mode}")

    ## INITIALIZATION
    text = load_text(source_files[lang], dir)
    ptgs = load_ptg_text(text)

    ## TAKE FIRST K WORDS
    ptgs = ptgs[:K]

    words_and_pos = load_word_and_pos(ptgs)

    if (mode == "words"):
        words, pos = separate_words_and_pos(words_and_pos)
    elif (mode == "pos"):
        pos, words = separate_words_and_pos(words_and_pos)
    else:
        raise ValueError("Invalid mode")
    
    N = total_number_of_words(words)
    C = total_number_of_unique_words(words)

    unique_words = get_unique_words(words)

    classes = arange_classes(unique_words, MIN_FREQ, get_word_counts(words))
    class_to_word = create_class_to_word_mapping(unique_words)
    word_to_class = create_word_to_class_mapping(unique_words)

    bigram_counts = get_bigram_counts(words)
    c_bi = create_bigram_matrix(unique_words, bigram_counts, word_to_class)
    c_l = create_unigram_matrix(unique_words, get_left_unigram_counts(bigram_counts), word_to_class)
    c_r = create_unigram_matrix(unique_words, get_right_unigram_counts(bigram_counts), word_to_class)

    Q, S = precompute_q_and_s(classes, c_bi, c_l, c_r, C, N)

    if verbose:
        print(f"Sum of the Q: {np.sum(Q)}")

    counts_current = (c_bi, c_l, c_r)

    ## MAIN LOOP
    L = init_l(Q, S, N, C, counts_current, classes)

    history = []
    human_readable_history = []

    while len(classes) > LIMIT:
        ## perform the update step

        ## find the best pair to merge
        best_pair, minimal_loss  = find_best_pair(L, classes)

        ## log
        if verbose:
            print(f"{class_to_word[best_pair[1]]} into {class_to_word[best_pair[0]]}, loss: {minimal_loss}")

        ## ajdust the counts
        counts_future = mergeCounts(counts_current, best_pair[0], best_pair[1])

        ## update the loss table, pick the best pair to merge
        L, S = update(L, S, counts_current, counts_future, classes, N, best_pair[0], best_pair[1])

        ## update the history
        history.append(best_pair)
        human_readable_history.append((class_to_word[best_pair[0]].copy(), class_to_word[best_pair[1]].copy()))

        ## update the class_to_word mapping
        class_to_word[best_pair[0]] += class_to_word[best_pair[1]]
        class_to_word.pop(best_pair[1])

        ## update the word_to_class mapping
        for word, cls in word_to_class.items():
            if cls == best_pair[1]:
                word_to_class[word] = best_pair[0]
        
        ## update the classes
        classes = np.delete(classes, np.where(classes == best_pair[1]))

        ## update the counts
        counts_current = counts_future

    if verbose:
        for i, (cls_one, cls_two) in enumerate(human_readable_history):
            print(f"{i+1}: {cls_two} into {cls_one}")
    
    return [class_to_word[cls] for cls in classes], human_readable_history
In [14]:
for mode in ["words", "pos"]:
    if mode == "words":
        LIMIT = 15
        MIN_FREQ = 10
        K = 8000
    elif mode == "pos":
        LIMIT = 10
        MIN_FREQ = 5
        K = 200000

    for lang in ["en", "cz"]:  

        print(f"""Language: {lang}.
Merging {mode} into {LIMIT} classes.
Considering first {K} words with min. freq. {MIN_FREQ}...\n""")
        
        classes, history_of_merges = execute_merging(lang, mode, LIMIT, MIN_FREQ, K, verbose=True)

        for i, cls in enumerate(classes):
            print(f"Class {i+1}: {cls}")
        print("\n")

        ## save history to a file
        with open(f"../results/{lang}_{mode}_{LIMIT}_history.txt", "w", encoding="utf-8") as f:
            for cls_one, cls_two in history_of_merges:
                f.write(f"{cls_two} into {cls_one}\n")
Language: en.
Merging words into 15 classes.
Considering first 8000 words with min. freq. 10...

Merging language: en, mode: words
Sum of the Q: 4.995892538589587
['case'] into ['subject'], loss: 0.0021965665335756945
['may'] into ['cannot'], loss: 0.002669139511099379
['structure'] into ['individuals'], loss: 0.0026748091526128705
['It'] into ['there'], loss: 0.003479400370452433
['even'] into ['less'], loss: 0.0036556390622295674
['variation'] into ['nature'], loss: 0.0036909506202147946
['see'] into ['shall'], loss: 0.0038977958561413565
['slight'] into ['short'], loss: 0.0039056390622295677
['distinct'] into ['certain'], loss: 0.003992156793911794
['manner'] into ['such'], loss: 0.004241409281117521
['must'] into ['can'], loss: 0.004262927799019989
['state'] into ['subject', 'case'], loss: 0.0042766639487693235
['differ'] into ['what'], loss: 0.0042987949406953965
[')'] into ['cases'], loss: 0.004456435556800399
['me'] into ['only'], loss: 0.00447802686684627
['how'] into ['nearly'], loss: 0.004540204221812921
['domesticated'] into ['domestic'], loss: 0.004563547053784825
['races'] into ['varieties'], loss: 0.004613365915588918
['if'] into ['when'], loss: 0.004762832413873117
['than'] into ['('], loss: 0.0048728052253528276
['do'] into ['believe'], loss: 0.0049492035054291604
['great'] into ['my'], loss: 0.005125548031345392
['could'] into ['will'], loss: 0.005600648386150989
['plants'] into ['facts'], loss: 0.0056896399601957266
['between'] into [':'], loss: 0.005755267783733755
['its'] into ['different'], loss: 0.005811278124459137
['conditions'] into ['nature', 'variation'], loss: 0.005854956030029314
['In'] into ['The'], loss: 0.005975681770144466
['long'] into ['short', 'slight'], loss: 0.0062716868916984116
['these'] into ['each'], loss: 0.0063407179814485605
['would'] into ['cannot', 'may'], loss: 0.006573989711723835
['most'] into ['individuals', 'structure'], loss: 0.006710581835738694
['very'] into ['any'], loss: 0.006827542130200615
['so'] into ['often'], loss: 0.007051575444189467
['we'] into ['they'], loss: 0.007132892772609472
['breeds'] into ['subject', 'case', 'state'], loss: 0.007212068207649337
['wild'] into ['certain', 'distinct'], loss: 0.007777689674487167
['less', 'even'] into ['such', 'manner'], loss: 0.008128809562529237
['it'] into ['there', 'It'], loss: 0.00814083806206068
['an'] into ['much'], loss: 0.008244846816464119
['can', 'must'] into ['cannot', 'may', 'would'], loss: 0.00836379129755873
['animals'] into ['varieties', 'races'], loss: 0.008558816507676306
['many'] into ['several'], loss: 0.008909921301797073
['all'] into ['what', 'differ'], loss: 0.009227791157599648
['only', 'me'] into ['believe', 'do'], loss: 0.009411342758844692
['but'] into ['(', 'than'], loss: 0.00956669905415268
['cases', ')'] into ['nearly', 'how'], loss: 0.00965124016553962
['other'] into ['one'], loss: 0.009719360382467134
['at'] into [':', 'between'], loss: 0.01028631175693167
['some'] into ['their'], loss: 0.010690608987174686
['more'] into ['shall', 'see'], loss: 0.011028134299625523
['as'] into ['when', 'if'], loss: 0.011158014327088295
['nature', 'variation', 'conditions'] into ['individuals', 'structure', 'most'], loss: 0.01133854228576027
['different', 'its'] into ['much', 'an'], loss: 0.012094766529275517
['my', 'great'] into ['short', 'slight', 'long'], loss: 0.012354424064353053
['for'] into ['under'], loss: 0.012517483448205951
['each', 'these'] into ['their', 'some'], loss: 0.012825038255811619
['species'] into ['varieties', 'races', 'animals'], loss: 0.012974398391393102
['will', 'could'] into ['cannot', 'may', 'would', 'can', 'must'], loss: 0.012986465443803719
['our'] into ['several', 'many'], loss: 0.01348182767925122
['any', 'very'] into ['what', 'differ', 'all'], loss: 0.013459339217764085
['facts', 'plants'] into ['such', 'manner', 'less', 'even'], loss: 0.013745183113729156
['with'] into ['often', 'so'], loss: 0.01406037032451422
['certain', 'distinct', 'wild'] into ['one', 'other'], loss: 0.015344402398685454
['I'] into ['they', 'we'], loss: 0.01563365339577806
['has'] into ['not'], loss: 0.015973041777880122
['same'] into ['subject', 'case', 'state', 'breeds'], loss: 0.01649447587886521
['domestic', 'domesticated'] into ['one', 'other', 'certain', 'distinct', 'wild'], loss: 0.016525149034757675
['nearly', 'how', 'cases', ')'] into ['shall', 'see', 'more'], loss: 0.017458719916070224
['by'] into ['on'], loss: 0.017699186641382918
['which'] into ['this'], loss: 0.017779702329765442
['believe', 'do', 'only', 'me'] into ['(', 'than', 'but'], loss: 0.01828753914667661
['been'] into [':', 'between', 'at'], loss: 0.020305525538123863
['or'] into ['under', 'for'], loss: 0.022322604123214074
['from'] into ['often', 'so', 'with'], loss: 0.022452873285990185
['is'] into ['are'], loss: 0.02330965397012181
['varieties', 'races', 'animals', 'species'] into ['such', 'manner', 'less', 'even', 'facts', 'plants'], loss: 0.02332631429369478
['The', 'In'] into ['there', 'It', 'it'], loss: 0.023791644897939068
['have'] into ['not', 'has'], loss: 0.024063043420479536
['much', 'an', 'different', 'its'] into ['their', 'some', 'each', 'these'], loss: 0.02505109884743363
['short', 'slight', 'long', 'my', 'great'] into ['what', 'differ', 'all', 'any', 'very'], loss: 0.025710589246418065
['they', 'we', 'I'] into ['there', 'It', 'it', 'The', 'In'], loss: 0.0265967329747308
['that'] into ['when', 'if', 'as'], loss: 0.0266400300416021
['a'] into ['several', 'many', 'our'], loss: 0.029531699827928234
['be'] into ['(', 'than', 'but', 'believe', 'do', 'only', 'me'], loss: 0.03284669379207519
['shall', 'see', 'more', 'nearly', 'how', 'cases', ')'] into ['what', 'differ', 'all', 'any', 'very', 'short', 'slight', 'long', 'my', 'great'], loss: 0.033760944016454436
[':', 'between', 'at', 'been'] into ['often', 'so', 'with', 'from'], loss: 0.03516458858913665
['in'] into ['on', 'by'], loss: 0.03601025146728219
['subject', 'case', 'state', 'breeds', 'same'] into ['individuals', 'structure', 'most', 'nature', 'variation', 'conditions'], loss: 0.03974470674006333
['this', 'which'] into ['their', 'some', 'each', 'these', 'much', 'an', 'different', 'its'], loss: 0.039877009077549604
['such', 'manner', 'less', 'even', 'facts', 'plants', 'varieties', 'races', 'animals', 'species'] into ['individuals', 'structure', 'most', 'nature', 'variation', 'conditions', 'subject', 'case', 'state', 'breeds', 'same'], loss: 0.04000412379498797
[';'] into ['cannot', 'may', 'would', 'can', 'must', 'will', 'could'], loss: 0.040100532347277756
['under', 'for', 'or'] into ['are', 'is'], loss: 0.042847519208385344
['not', 'has', 'have'] into ['(', 'than', 'but', 'believe', 'do', 'only', 'me', 'be'], loss: 0.04414943404587341
['one', 'other', 'certain', 'distinct', 'wild', 'domestic', 'domesticated'] into ['several', 'many', 'our', 'a'], loss: 0.04510962278374156
['and'] into ['when', 'if', 'as', 'that'], loss: 0.053149087467073634
['what', 'differ', 'all', 'any', 'very', 'short', 'slight', 'long', 'my', 'great', 'shall', 'see', 'more', 'nearly', 'how', 'cases', ')'] into ['their', 'some', 'each', 'these', 'much', 'an', 'different', 'its', 'this', 'which'], loss: 0.06085473265433029
1: ['case'] into ['subject']
2: ['may'] into ['cannot']
3: ['structure'] into ['individuals']
4: ['It'] into ['there']
5: ['even'] into ['less']
6: ['variation'] into ['nature']
7: ['see'] into ['shall']
8: ['slight'] into ['short']
9: ['distinct'] into ['certain']
10: ['manner'] into ['such']
11: ['must'] into ['can']
12: ['state'] into ['subject', 'case']
13: ['differ'] into ['what']
14: [')'] into ['cases']
15: ['me'] into ['only']
16: ['how'] into ['nearly']
17: ['domesticated'] into ['domestic']
18: ['races'] into ['varieties']
19: ['if'] into ['when']
20: ['than'] into ['(']
21: ['do'] into ['believe']
22: ['great'] into ['my']
23: ['could'] into ['will']
24: ['plants'] into ['facts']
25: ['between'] into [':']
26: ['its'] into ['different']
27: ['conditions'] into ['nature', 'variation']
28: ['In'] into ['The']
29: ['long'] into ['short', 'slight']
30: ['these'] into ['each']
31: ['would'] into ['cannot', 'may']
32: ['most'] into ['individuals', 'structure']
33: ['very'] into ['any']
34: ['so'] into ['often']
35: ['we'] into ['they']
36: ['breeds'] into ['subject', 'case', 'state']
37: ['wild'] into ['certain', 'distinct']
38: ['less', 'even'] into ['such', 'manner']
39: ['it'] into ['there', 'It']
40: ['an'] into ['much']
41: ['can', 'must'] into ['cannot', 'may', 'would']
42: ['animals'] into ['varieties', 'races']
43: ['many'] into ['several']
44: ['all'] into ['what', 'differ']
45: ['only', 'me'] into ['believe', 'do']
46: ['but'] into ['(', 'than']
47: ['cases', ')'] into ['nearly', 'how']
48: ['other'] into ['one']
49: ['at'] into [':', 'between']
50: ['some'] into ['their']
51: ['more'] into ['shall', 'see']
52: ['as'] into ['when', 'if']
53: ['nature', 'variation', 'conditions'] into ['individuals', 'structure', 'most']
54: ['different', 'its'] into ['much', 'an']
55: ['my', 'great'] into ['short', 'slight', 'long']
56: ['for'] into ['under']
57: ['each', 'these'] into ['their', 'some']
58: ['species'] into ['varieties', 'races', 'animals']
59: ['will', 'could'] into ['cannot', 'may', 'would', 'can', 'must']
60: ['our'] into ['several', 'many']
61: ['any', 'very'] into ['what', 'differ', 'all']
62: ['facts', 'plants'] into ['such', 'manner', 'less', 'even']
63: ['with'] into ['often', 'so']
64: ['certain', 'distinct', 'wild'] into ['one', 'other']
65: ['I'] into ['they', 'we']
66: ['has'] into ['not']
67: ['same'] into ['subject', 'case', 'state', 'breeds']
68: ['domestic', 'domesticated'] into ['one', 'other', 'certain', 'distinct', 'wild']
69: ['nearly', 'how', 'cases', ')'] into ['shall', 'see', 'more']
70: ['by'] into ['on']
71: ['which'] into ['this']
72: ['believe', 'do', 'only', 'me'] into ['(', 'than', 'but']
73: ['been'] into [':', 'between', 'at']
74: ['or'] into ['under', 'for']
75: ['from'] into ['often', 'so', 'with']
76: ['is'] into ['are']
77: ['varieties', 'races', 'animals', 'species'] into ['such', 'manner', 'less', 'even', 'facts', 'plants']
78: ['The', 'In'] into ['there', 'It', 'it']
79: ['have'] into ['not', 'has']
80: ['much', 'an', 'different', 'its'] into ['their', 'some', 'each', 'these']
81: ['short', 'slight', 'long', 'my', 'great'] into ['what', 'differ', 'all', 'any', 'very']
82: ['they', 'we', 'I'] into ['there', 'It', 'it', 'The', 'In']
83: ['that'] into ['when', 'if', 'as']
84: ['a'] into ['several', 'many', 'our']
85: ['be'] into ['(', 'than', 'but', 'believe', 'do', 'only', 'me']
86: ['shall', 'see', 'more', 'nearly', 'how', 'cases', ')'] into ['what', 'differ', 'all', 'any', 'very', 'short', 'slight', 'long', 'my', 'great']
87: [':', 'between', 'at', 'been'] into ['often', 'so', 'with', 'from']
88: ['in'] into ['on', 'by']
89: ['subject', 'case', 'state', 'breeds', 'same'] into ['individuals', 'structure', 'most', 'nature', 'variation', 'conditions']
90: ['this', 'which'] into ['their', 'some', 'each', 'these', 'much', 'an', 'different', 'its']
91: ['such', 'manner', 'less', 'even', 'facts', 'plants', 'varieties', 'races', 'animals', 'species'] into ['individuals', 'structure', 'most', 'nature', 'variation', 'conditions', 'subject', 'case', 'state', 'breeds', 'same']
92: [';'] into ['cannot', 'may', 'would', 'can', 'must', 'will', 'could']
93: ['under', 'for', 'or'] into ['are', 'is']
94: ['not', 'has', 'have'] into ['(', 'than', 'but', 'believe', 'do', 'only', 'me', 'be']
95: ['one', 'other', 'certain', 'distinct', 'wild', 'domestic', 'domesticated'] into ['several', 'many', 'our', 'a']
96: ['and'] into ['when', 'if', 'as', 'that']
97: ['what', 'differ', 'all', 'any', 'very', 'short', 'slight', 'long', 'my', 'great', 'shall', 'see', 'more', 'nearly', 'how', 'cases', ')'] into ['their', 'some', 'each', 'these', 'much', 'an', 'different', 'its', 'this', 'which']
Class 1: ['cannot', 'may', 'would', 'can', 'must', 'will', 'could', ';']
Class 2: ['individuals', 'structure', 'most', 'nature', 'variation', 'conditions', 'subject', 'case', 'state', 'breeds', 'same', 'such', 'manner', 'less', 'even', 'facts', 'plants', 'varieties', 'races', 'animals', 'species']
Class 3: ['when', 'if', 'as', 'that', 'and']
Class 4: ['are', 'is', 'under', 'for', 'or']
Class 5: ['their', 'some', 'each', 'these', 'much', 'an', 'different', 'its', 'this', 'which', 'what', 'differ', 'all', 'any', 'very', 'short', 'slight', 'long', 'my', 'great', 'shall', 'see', 'more', 'nearly', 'how', 'cases', ')']
Class 6: ['several', 'many', 'our', 'a', 'one', 'other', 'certain', 'distinct', 'wild', 'domestic', 'domesticated']
Class 7: ['on', 'by', 'in']
Class 8: ['often', 'so', 'with', 'from', ':', 'between', 'at', 'been']
Class 9: ['of']
Class 10: ['there', 'It', 'it', 'The', 'In', 'they', 'we', 'I']
Class 11: ['(', 'than', 'but', 'believe', 'do', 'only', 'me', 'be', 'not', 'has', 'have']
Class 12: ['to']
Class 13: ['the']
Class 14: ['.']
Class 15: [',']


Language: cz.
Merging words into 15 classes.
Considering first 8000 words with min. freq. 10...

Merging language: cz, mode: words
Sum of the Q: 7.557199800862522
['listopadu'] into ['OKD'], loss: 0.003082884270599922
['který'] into ['které'], loss: 0.0033734316421669547
['státu'] into ['J'], loss: 0.004024966367527468
['bude'] into ['musí'], loss: 0.0044216001660482285
['aby'] into ['ale'], loss: 0.004603782679326707
['nás'] into ['bylo'], loss: 0.004647048343347073
['si'] into ['pouze'], loss: 0.004991409281117513
['už'] into ['mezi'], loss: 0.005000000000000001
['&slash;'] into ['NATO'], loss: 0.00556127812445914
['ze'] into ['jeho'], loss: 0.00556426990688624
['však'] into ['být'], loss: 0.005742784599970021
['byl'] into ['jsou'], loss: 0.005749999999999991
['před'] into ['?'], loss: 0.0059999999999999915
['budou'] into ['jako'], loss: 0.0062500000000000056
['při'] into ['zákona'], loss: 0.006899419316170906
['V'] into ['Na'], loss: 0.007058395693542378
['OKD', 'listopadu'] into ['J', 'státu'], loss: 0.007106558562877181
['za'] into ['u'], loss: 0.007555812533852423
['včera'] into ['bylo', 'nás'], loss: 0.008115293978715554
['ČSFR'] into ['od'], loss: 0.00872995224880984
['že'] into ['které', 'který'], loss: 0.00919518545632083
['pouze', 'si'] into ['mezi', 'už'], loss: 0.010057083865413086
['pro'] into ['po'], loss: 0.010249999999999981
['musí', 'bude'] into ['?', 'před'], loss: 0.010746796771587318
['ale', 'aby'] into ['NATO', '&slash;'], loss: 0.011340652565987524
['k'] into ['jeho', 'ze'], loss: 0.011537365052878796
['být', 'však'] into ['jako', 'budou'], loss: 0.011652102818357001
['jsou', 'byl'] into ['zákona', 'při'], loss: 0.012310500446943484
['ve'] into ['to'], loss: 0.013698431273825466
['by'] into ['bylo', 'nás', 'včera'], loss: 0.014746339632353445
['Na', 'V'] into ['J', 'státu', 'OKD', 'listopadu'], loss: 0.015035659754277397
['do'] into ['u', 'za'], loss: 0.015907961185191034
['je'] into ['i'], loss: 0.016172407347170403
[':'] into ['od', 'ČSFR'], loss: 0.017749999999999985
['které', 'který', 'že'] into ['NATO', '&slash;', 'ale', 'aby'], loss: 0.01939815642260082
['mezi', 'už', 'pouze', 'si'] into ['po', 'pro'], loss: 0.01984223927847203
[')'] into ['z'], loss: 0.02215279953493207
['('] into ['?', 'před', 'musí', 'bude'], loss: 0.022553603635574095
['jako', 'budou', 'být', 'však'] into ['to', 've'], loss: 0.022948595627144276
['jeho', 'ze', 'k'] into ['zákona', 'při', 'jsou', 'byl'], loss: 0.02443383437337742
['s'] into ['"'], loss: 0.028059336263857856
['-'] into ['bylo', 'nás', 'včera', 'by'], loss: 0.03350440283263173
['od', 'ČSFR', ':'] into ['z', ')'], loss: 0.03382149888328662
['o'] into ['u', 'za', 'do'], loss: 0.03490333247494361
['J', 'státu', 'OKD', 'listopadu', 'Na', 'V'] into ['i', 'je'], loss: 0.0369042623484128
['po', 'pro', 'mezi', 'už', 'pouze', 'si'] into ['zákona', 'při', 'jsou', 'byl', 'jeho', 'ze', 'k'], loss: 0.04037380577853852
1: ['listopadu'] into ['OKD']
2: ['který'] into ['které']
3: ['státu'] into ['J']
4: ['bude'] into ['musí']
5: ['aby'] into ['ale']
6: ['nás'] into ['bylo']
7: ['si'] into ['pouze']
8: ['už'] into ['mezi']
9: ['&slash;'] into ['NATO']
10: ['ze'] into ['jeho']
11: ['však'] into ['být']
12: ['byl'] into ['jsou']
13: ['před'] into ['?']
14: ['budou'] into ['jako']
15: ['při'] into ['zákona']
16: ['V'] into ['Na']
17: ['OKD', 'listopadu'] into ['J', 'státu']
18: ['za'] into ['u']
19: ['včera'] into ['bylo', 'nás']
20: ['ČSFR'] into ['od']
21: ['že'] into ['které', 'který']
22: ['pouze', 'si'] into ['mezi', 'už']
23: ['pro'] into ['po']
24: ['musí', 'bude'] into ['?', 'před']
25: ['ale', 'aby'] into ['NATO', '&slash;']
26: ['k'] into ['jeho', 'ze']
27: ['být', 'však'] into ['jako', 'budou']
28: ['jsou', 'byl'] into ['zákona', 'při']
29: ['ve'] into ['to']
30: ['by'] into ['bylo', 'nás', 'včera']
31: ['Na', 'V'] into ['J', 'státu', 'OKD', 'listopadu']
32: ['do'] into ['u', 'za']
33: ['je'] into ['i']
34: [':'] into ['od', 'ČSFR']
35: ['které', 'který', 'že'] into ['NATO', '&slash;', 'ale', 'aby']
36: ['mezi', 'už', 'pouze', 'si'] into ['po', 'pro']
37: [')'] into ['z']
38: ['('] into ['?', 'před', 'musí', 'bude']
39: ['jako', 'budou', 'být', 'však'] into ['to', 've']
40: ['jeho', 'ze', 'k'] into ['zákona', 'při', 'jsou', 'byl']
41: ['s'] into ['"']
42: ['-'] into ['bylo', 'nás', 'včera', 'by']
43: ['od', 'ČSFR', ':'] into ['z', ')']
44: ['o'] into ['u', 'za', 'do']
45: ['J', 'státu', 'OKD', 'listopadu', 'Na', 'V'] into ['i', 'je']
46: ['po', 'pro', 'mezi', 'už', 'pouze', 'si'] into ['zákona', 'při', 'jsou', 'byl', 'jeho', 'ze', 'k']
Class 1: ['bylo', 'nás', 'včera', 'by', '-']
Class 2: ['u', 'za', 'do', 'o']
Class 3: ['z', ')', 'od', 'ČSFR', ':']
Class 4: ['se']
Class 5: ['?', 'před', 'musí', 'bude', '(']
Class 6: ['a']
Class 7: ['zákona', 'při', 'jsou', 'byl', 'jeho', 'ze', 'k', 'po', 'pro', 'mezi', 'už', 'pouze', 'si']
Class 8: ['i', 'je', 'J', 'státu', 'OKD', 'listopadu', 'Na', 'V']
Class 9: ['to', 've', 'jako', 'budou', 'být', 'však']
Class 10: ['NATO', '&slash;', 'ale', 'aby', 'které', 'který', 'že']
Class 11: ['na']
Class 12: ['.']
Class 13: ['"', 's']
Class 14: ['v']
Class 15: [',']


Language: en.
Merging pos into 10 classes.
Considering first 200000 words with min. freq. 5...

Merging language: en, mode: pos
Sum of the Q: 0.8825168031143319
['WP$'] into ['RBR'], loss: 0.00021477719197586455
['RBR', 'WP$'] into ['JJR'], loss: 0.00038471950695165237
['FW'] into ['NNPS'], loss: 0.0009191375937520887
['NNPS', 'FW'] into ['NNP'], loss: 0.0008363385508621645
['('] into ['.'], loss: 0.0010989642349956289
['EX'] into ['PRP'], loss: 0.0011060376076082457
['"'] into ['WP'], loss: 0.0014244387177662697
['JJS'] into ['JJ'], loss: 0.001729428258254846
['RBS'] into ['JJR', 'RBR', 'WP$'], loss: 0.002114096705532564
['WRB'] into ['WP', '"'], loss: 0.002200261876482527
['SYM'] into ['VBD'], loss: 0.002760619388733474
['DT'] into ['PRP$'], loss: 0.0027498495389207487
['CD'] into ['JJ', 'JJS'], loss: 0.00346420026378344
['WDT'] into ['WP', '"', 'WRB'], loss: 0.005802251430491217
[':'] into [','], loss: 0.005994340401075265
['VBN'] into ['VBD', 'SYM'], loss: 0.007252006756299058
['VBP'] into ['VBZ'], loss: 0.007065800672448878
['VBG'] into ['JJR', 'RBR', 'WP$', 'RBS'], loss: 0.008937882258500894
['NNS'] into ['NN'], loss: 0.008985748580892529
['.', '('] into [',', ':'], loss: 0.014538933148508404
['JJR', 'RBR', 'WP$', 'RBS', 'VBG'] into ['NNP', 'NNPS', 'FW'], loss: 0.016113465288245567
['WP', '"', 'WRB', 'WDT'] into ['NNP', 'NNPS', 'FW', 'JJR', 'RBR', 'WP$', 'RBS', 'VBG'], loss: 0.01833150630778286
['MD'] into ['TO'], loss: 0.026436117447222685
['RB'] into ['NNP', 'NNPS', 'FW', 'JJR', 'RBR', 'WP$', 'RBS', 'VBG', 'WP', '"', 'WRB', 'WDT'], loss: 0.0276992597896548
['VB'] into ['VBZ', 'VBP'], loss: 0.030610132237013896
['PRP', 'EX'] into ['NNP', 'NNPS', 'FW', 'JJR', 'RBR', 'WP$', 'RBS', 'VBG', 'WP', '"', 'WRB', 'WDT', 'RB'], loss: 0.032981376181005884
1: ['WP$'] into ['RBR']
2: ['RBR', 'WP$'] into ['JJR']
3: ['FW'] into ['NNPS']
4: ['NNPS', 'FW'] into ['NNP']
5: ['('] into ['.']
6: ['EX'] into ['PRP']
7: ['"'] into ['WP']
8: ['JJS'] into ['JJ']
9: ['RBS'] into ['JJR', 'RBR', 'WP$']
10: ['WRB'] into ['WP', '"']
11: ['SYM'] into ['VBD']
12: ['DT'] into ['PRP$']
13: ['CD'] into ['JJ', 'JJS']
14: ['WDT'] into ['WP', '"', 'WRB']
15: [':'] into [',']
16: ['VBN'] into ['VBD', 'SYM']
17: ['VBP'] into ['VBZ']
18: ['VBG'] into ['JJR', 'RBR', 'WP$', 'RBS']
19: ['NNS'] into ['NN']
20: ['.', '('] into [',', ':']
21: ['JJR', 'RBR', 'WP$', 'RBS', 'VBG'] into ['NNP', 'NNPS', 'FW']
22: ['WP', '"', 'WRB', 'WDT'] into ['NNP', 'NNPS', 'FW', 'JJR', 'RBR', 'WP$', 'RBS', 'VBG']
23: ['MD'] into ['TO']
24: ['RB'] into ['NNP', 'NNPS', 'FW', 'JJR', 'RBR', 'WP$', 'RBS', 'VBG', 'WP', '"', 'WRB', 'WDT']
25: ['VB'] into ['VBZ', 'VBP']
26: ['PRP', 'EX'] into ['NNP', 'NNPS', 'FW', 'JJR', 'RBR', 'WP$', 'RBS', 'VBG', 'WP', '"', 'WRB', 'WDT', 'RB']
Class 1: ['VBZ', 'VBP', 'VB']
Class 2: ['TO', 'MD']
Class 3: ['NNP', 'NNPS', 'FW', 'JJR', 'RBR', 'WP$', 'RBS', 'VBG', 'WP', '"', 'WRB', 'WDT', 'RB', 'PRP', 'EX']
Class 4: ['VBD', 'SYM', 'VBN']
Class 5: ['PRP$', 'DT']
Class 6: ['CC']
Class 7: ['IN']
Class 8: [',', ':', '.', '(']
Class 9: ['JJ', 'JJS', 'CD']
Class 10: ['NN', 'NNS']


Language: cz.
Merging pos into 10 classes.
Considering first 200000 words with min. freq. 5...

Merging language: cz, mode: pos
Sum of the Q: 1.6931364783689442
['AAFS6----3A----'] into ['PDFS6----------'], loss: 6.602429515910335e-06
['AAFS6----1N----'] into ['PSFS6-P1-------'], loss: 1.7397219203907138e-05
['PSZS6-P1-------'] into ['CrIS6----------'], loss: 3.127729972883606e-05
['PLXP6----------'] into ['AGFP6-----A----'], loss: 3.4311009749392676e-05
['AAIP6----3A----'] into ['AAIP6----1A----'], loss: 3.823113425607377e-05
['CrFS6----------'] into ['PSFS6-P1-------', 'AAFS6----1N----'], loss: 4.086610136210889e-05
['AAIS6----1N----'] into ['PSZS6FS3-------'], loss: 4.474247407439927e-05
['PZXP6----------'] into ['Ca--6----------'], loss: 4.55535946066888e-05
['CrNS6----------'] into ['AUNS6M---------'], loss: 4.7869141299980694e-05
['AUFS7M---------'] into ['AAFS7----1N----'], loss: 4.888926530050269e-05
['PJZS3----------'] into ['P4MS4----------'], loss: 5.452478119021985e-05
['J^------------8'] into ['VB-S---2P-NA---'], loss: 5.616887333118688e-05
['AGIS7-----A----'] into ['AAIS7----3A----'], loss: 5.753089974986855e-05
['ClIS4----------'] into ['CrIS4----------'], loss: 5.805235962353457e-05
['ClZS6----------'] into ['PSZS6FS3-------', 'AAIS6----1N----'], loss: 5.823773739519248e-05
['AAFP7----1A----'] into ['AAFP7----2A----'], loss: 6.18338460113034e-05
['AGFS6-----A----'] into ['P4FS6----------'], loss: 6.295422134917569e-05
['PSHS1-S1-------'] into ['PSHS1-P1-------'], loss: 6.327731653869742e-05
['AANS7----1A----'] into ['AANS7----1N----'], loss: 6.757107827307703e-05
['CrFS2----------'] into ['AAFS2----2A----'], loss: 6.887488872184201e-05
['AGFS7-----A----'] into ['AAFS7----3A----'], loss: 7.180820790410845e-05
['PZFS6----------'] into ['ClFS6----------'], loss: 7.216505985532242e-05
['CrNS2----------'] into ['AUNS2M---------'], loss: 7.242132536893354e-05
['Cn-P6----------'] into ['PSXP6-P1-------'], loss: 7.355760465339303e-05
['PZFS1----------'] into ['PWFS1----------'], loss: 7.411536904791975e-05
['AUFS6M---------'] into ['PSFS6-P1-------', 'AAFS6----1N----', 'CrFS6----------'], loss: 7.486445036578894e-05
['AAIP2----2A----'] into ['AAIP2----1N----'], loss: 7.568413021558822e-05
['P5FS2--3-------'] into ['P5ZS2--3------1'], loss: 7.613910461431246e-05
['P4FS7----------'] into ['P9ZS7----------'], loss: 7.661663906016161e-05
['P9XP6----------'] into ['PQ--6----------'], loss: 7.861325782343312e-05
['PSHP1-P2-------'] into ['CrIP1----------'], loss: 7.8978285247296e-05
['NNNSX-----A----'] into ['P6-X6----------'], loss: 7.983773783657019e-05
['AAIP4----1N----'] into ['PZYP4----------'], loss: 8.11819960232483e-05
['AAIS2----2A----'] into ['AAIS2----3A----'], loss: 8.178547770503939e-05
['NNISX-----A----'] into ['AAIS1----1A---6'], loss: 8.287676760590035e-05
['PKM-4----------'] into ['ClHP4----------'], loss: 8.287676760590046e-05
['AUMS1M---------'] into ['AAMS1----1N----'], loss: 8.506559646197513e-05
['PQ--2----------'] into ['PKM-3----------'], loss: 8.659745483682976e-05
['PLIP1----------'] into ['PDIP1----------'], loss: 8.678575894221813e-05
['AGIS4-----A----'] into ['AAIS4----3A----'], loss: 8.766761376346045e-05
['PWFS4----------'] into ['PZFS4----------'], loss: 8.813236019583441e-05
['ACQW------A----'] into ['VB-S---1F-AA---'], loss: 8.839324729066904e-05
['AGNP2-----A----'] into ['CrNP2----------'], loss: 8.936122710717827e-05
['AAMS7----3A----'] into ['AAMS7----1A----'], loss: 8.953842816170152e-05
['CrIS6----------', 'PSZS6-P1-------'] into ['PSZS6FS3-------', 'AAIS6----1N----', 'ClZS6----------'], loss: 9.08189867067739e-05
['AAMP1----2A----'] into ['PSMP1-P1-------'], loss: 9.148631924391104e-05
['PJNS1----------'] into ['PJYS1----------'], loss: 9.363837013045997e-05
['AAIP1----3A----'] into ['AAIP1----1N----'], loss: 9.366961442515615e-05
['CrIS7----------'] into ['VeYS------A----'], loss: 9.594545399532067e-05
['PDFS7----------'] into ['AAFS7----1N----', 'AUFS7M---------'], loss: 9.61859381284748e-05
['P8XP6----------'] into ['AGFP6-----A----', 'PLXP6----------'], loss: 9.74916285957127e-05
['AUIS1M---------'] into ['PWYS1----------'], loss: 9.822407687161449e-05
['P5FS3--3-------'] into ['P5ZS3--3-------'], loss: 9.968393113167777e-05
['AAIP7----3A----'] into ['NNNP1-----A---2'], loss: 0.00010098784914196288
['ClMS4----------'] into ['P8MS4----------'], loss: 0.00010114378539174149
['ACTP------A----'] into ['P6-X3----------'], loss: 0.00010199924914586678
['NNFS2-----N----'] into ['AGNP6-----A----'], loss: 0.00010403738788900647
['AGFP1-----A----'] into ['CrFP1----------'], loss: 0.00010452519610807613
['PP-P7--1-------'] into ['P6-X7----------'], loss: 0.00010570640120894628
['AAFP2----3A----'] into ['AAFP2----2A----'], loss: 0.00010632893777737457
['PSYS1-S1-------'] into ['PSYS1-P1-------'], loss: 0.00010707977018945014
['P4NS4----------'] into ['J,-S---1-------'], loss: 0.00010893409705288871
['Dg-------3A---1'] into ['Dg-------1A---1'], loss: 0.00011040711547118077
['AGFP7-----A----'] into ['P9ZS7----------', 'P4FS7----------'], loss: 0.00011078938183790214
['P5XP6--3-------'] into ['P5FS6--3-------'], loss: 0.00011096184852375226
['PJXP1----------'] into ['P4XP3----------'], loss: 0.00011219240798770428
['AAFS3----1A----'] into ['PDFS3----------'], loss: 0.00011356527916667278
['AGNS2-----A----'] into ['NNNS1-----N----'], loss: 0.00011362587118065118
['PLZS6----------'] into ['AUIS6M---------'], loss: 0.0001165455797007785
['CrNS4----------'] into ['AANS4----1N----'], loss: 0.00012024031872400246
['AUFS4M---------'] into ['AAFS4----3A----'], loss: 0.00012205921732702193
['AAFS4----1N----'] into ['P8FS4----------'], loss: 0.0001223441458035508
['P4YP4----------'] into ['P4MS4----------', 'PJZS3----------'], loss: 0.00012268510453151732
['PZIS4----------'] into ['AAIS4----1N----'], loss: 0.00012460674958518362
['P8XP7----------'] into ['Cn-P7----------'], loss: 0.00012469188836696757
['ClFS2----------'] into ['AAFS2----1N----'], loss: 0.00012495138800340901
['PDXP6----------'] into ['Ca--6----------', 'PZXP6----------'], loss: 0.00012616994047902124
['AAIS2----1N----'] into ['AUIS2M---------'], loss: 0.0001268898734153995
['VB-S---1P-NA--1'] into ['VB-P---1F-AA---'], loss: 0.00012745111073816972
['PP-P1--2-------'] into ['AAIP1----2A----'], loss: 0.00012783419647870598
['PSZS2-S1-------'] into ['PSZS2FS3-------'], loss: 0.00012820781931459574
['AAFP4----1N----'] into ['PSXP4-P1-------'], loss: 0.00013097032849878178
['AAMP1----3A----'] into ['AGMP1-----A----'], loss: 0.00013423749482528897
['PZZS7----------'] into ['PLNS4---------1'], loss: 0.00013462998512546947
['PZM-4----------'] into ['VB-S---1F-AA---', 'ACQW------A----'], loss: 0.00013492399981792128
['AAMS6----1A----'] into ['NNMS6-----A---1'], loss: 0.00013707499822178892
['Co-------------'] into ['Dg-------1A---3'], loss: 0.00013843708239592043
['AAFP1----3A----'] into ['PLFP1----------'], loss: 0.00013863727975556316
['P4XP2----------'] into ['P6-X2----------'], loss: 0.00013914136825358297
['AANS1----1N----'] into ['AANS1----2A----'], loss: 0.0001392613395586721
['PQ--7----------'] into ['J^------------1'], loss: 0.00014125222406912298
['PLXP3----------'] into ['ClXP3----------'], loss: 0.0001422319980771262
['NNFP1-----N----'] into ['NNMP6-----A----'], loss: 0.00014528698157553407
['J,-P---1-------'] into ['Vi-P---2--N----'], loss: 0.00014542949535726745
['AGIS2-----A----'] into ['ClZS3----------'], loss: 0.00014621545865437424
['Vi-S---2--A----'] into ['VB-S---2P-NA---', 'J^------------8'], loss: 0.00014673469019407348
['AAFP1----1N----'] into ['AAFP1----2A----'], loss: 0.00014721114610258853
['NNIP2-----A---1'] into ['NNFS1-----N----'], loss: 0.00014734916831885252
['P9XP2----------'] into ['P5ZS2--3------1', 'P5FS2--3-------'], loss: 0.00014861570198849744
['AAXXX----1A---8'] into ['A2--------A----'], loss: 0.00014953583876646656
['NNFP2-----A---1'] into ['NNIS5-----A----'], loss: 0.0001508327581453083
['CyFP1----------'] into ['AAIS1----1A---6', 'NNISX-----A----'], loss: 0.00015113982593553405
['AAMS1----2A----'] into ['CrMS1----------'], loss: 0.00015607721280773987
['P9ZS6----------'] into ['PQ--6----------', 'P9XP6----------'], loss: 0.00015896876411512426
['PDXP7----------'] into ['AAIS7----1N----'], loss: 0.00015906220039315863
['II-------------'] into ['NNFS5-----A----'], loss: 0.0001593807792310604
['C?--4----------'] into ['NNNS1-----N----', 'AGNS2-----A----'], loss: 0.00016584349122706005
['CrNP1----------'] into ['AANP1----1A----'], loss: 0.00016606226380311197
['CrIP1----------', 'PSHP1-P2-------'] into ['AAIP1----1N----', 'AAIP1----3A----'], loss: 0.0001665291446984363
['P8FP4---------1'] into ['PLFP4----------'], loss: 0.00016681379923999453
['VB-P---3F-NA---'] into ['VB-P---3P-NA--1'], loss: 0.0001692071897158036
['PLNS1---------1'] into ['ClNS1----------'], loss: 0.00016990451942833999
['PSZS1FS3-------'] into ['PZYS1----------'], loss: 0.00017087322714851526
['P6-X4----------'] into ['P5ZS4--3-------'], loss: 0.00017156485440534847
['NNFS2-----A---1'] into ['AGNP6-----A----', 'NNFS2-----N----'], loss: 0.0001720042265890269
['AAFS7----3A----', 'AGFS7-----A----'] into ['AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------'], loss: 0.0001743127751912309
['PDMS4----------'] into ['VB-P---2P-NA---'], loss: 0.00017478198242740326
['PSFS4-P1-------'] into ['PZFS4----------', 'PWFS4----------'], loss: 0.00017482692246999416
['CrFS4----------'] into ['AAFS4----3A----', 'AUFS4M---------'], loss: 0.00017495957586905505
['AAIS1----1N----'] into ['AAIS1----2A----'], loss: 0.00017617852748895392
['PSXP2-P1-------'] into ['AAFP2----2A----', 'AAFP2----3A----'], loss: 0.000177744110711148
['P5XP7--3-------'] into ['P5ZS7--3-------'], loss: 0.00017937870682232203
['PDIP4----------'] into ['PZYP4----------', 'AAIP4----1N----'], loss: 0.00017940732058412904
['P5ZS6--3-------'] into ['P9FS6----------'], loss: 0.00018073855084608896
['PPXP2--3-------'] into ['PPZS7--3-------'], loss: 0.00018144545932726906
['ClXP6----------'] into ['PSXP6-P1-------', 'Cn-P6----------'], loss: 0.0001816870020965788
['PP-P2--1-------'] into ['P5XP2--3-------'], loss: 0.0001823718156486508
['AGIP2-----A----'] into ['AAIP2----1N----', 'AAIP2----2A----'], loss: 0.00018289236305195052
['P4ZS6----------'] into ['P8ZS6----------'], loss: 0.00018317711043890897
['PZFP1----------'] into ['PSHP1-P1-------'], loss: 0.00018387965791479737
['PZMP1----------'] into ['PPMP1--3-------'], loss: 0.0001852795735940271
['AAMP2----1N----'] into ['AGMP2-----A----'], loss: 0.00018627294405387557
['PZFP4----------'] into ['PWFP4----------'], loss: 0.00018716701076192045
['AANS1----3A----'] into ['PLNS4----------'], loss: 0.00018821089164023763
['P5FS4--3-------'] into ['P5XP4--3-------'], loss: 0.00019170156619993791
['RR--X----------'] into ['RR--1----------'], loss: 0.00019638406637867683
['P4ZS7----------'] into ['PKM-3----------', 'PQ--2----------'], loss: 0.00019639055573759533
['ClFS4----------'] into ['AAFS4----2A----'], loss: 0.00019924067790500649
['P5ZS2--3-------'] into ['P6-X2----------', 'P4XP2----------'], loss: 0.0002010416221262818
['AANS2----1N----'] into ['AUNS2M---------', 'CrNS2----------'], loss: 0.00020122622909928673
['AUFS2M---------'] into ['AAFS2----2A----', 'CrFS2----------'], loss: 0.00020279016885173338
['Vf--------A---1'] into ['NNNP1-----A---2', 'AAIP7----3A----'], loss: 0.0002028166002103161
['PJYS1----------', 'PJNS1----------'] into ['J,-S---1-------', 'P4NS4----------'], loss: 0.00020373751003689602
['AAIS1----3A----'] into ['CrIS1----------'], loss: 0.00020412171537973812
['AGIS6-----A----'] into ['ClFS6----------', 'PZFS6----------'], loss: 0.0002060984192502088
['PSHS1-P1-------', 'PSHS1-S1-------'] into ['PWFS1----------', 'PZFS1----------'], loss: 0.0002064044606313823
['Vi-P---1--A----'] into ['Vi-P---2--A----'], loss: 0.00020643483801385787
['ClZS2----------'] into ['CrIS2----------'], loss: 0.0002084604997200577
['P6-X6----------', 'NNNSX-----A----'] into ['P5FS6--3-------', 'P5XP6--3-------'], loss: 0.000210025720627161
['ClXP7----------'] into ['Cn-P7----------', 'P8XP7----------'], loss: 0.00021167363742975514
['AGMS1-----A----'] into ['AAMS1----1N----', 'AUMS1M---------'], loss: 0.00021198644396826975
['PLFS1----------'] into ['P6-X3----------', 'ACTP------A----'], loss: 0.00021251103041784898
['PDMP4----------'] into ['PLNS4---------1', 'PZZS7----------'], loss: 0.0002144170404838699
['PLXP7----------'] into ['NNMP2-----A---1'], loss: 0.00021488390096222033
['PWYS1----------', 'AUIS1M---------'] into ['PZYS1----------', 'PSZS1FS3-------'], loss: 0.0002195148540697191
['NNIXX-----A---8'] into ['NNMXX-----A---8'], loss: 0.00021982108299129204
['P4FP1----------'] into ['P4IP1----------'], loss: 0.0002203726399170082
['P8FS6---------1'] into ['P4FS6----------', 'AGFS6-----A----'], loss: 0.00022066675710559936
['AUIS6M---------', 'PLZS6----------'] into ['P8ZS6----------', 'P4ZS6----------'], loss: 0.0002207468287509875
['AAIS7----3A----', 'AGIS7-----A----'] into ['AAIS7----1N----', 'PDXP7----------'], loss: 0.0002224065508947806
['PPNS4--3-------'] into ['P8ZS3----------'], loss: 0.0002246285940855873
['AGFS2-----A----'] into ['PSFS2-P1-------'], loss: 0.00023024442018551865
['AAFS6----1A----'] into ['PSFS6-P1-------', 'AAFS6----1N----', 'CrFS6----------', 'AUFS6M---------'], loss: 0.00023244046421189157
['P4FS4----------'] into ['P4MS4----------', 'PJZS3----------', 'P4YP4----------'], loss: 0.00023525723414344834
['AAMP3----1A----'] into ['P8XP3----------'], loss: 0.00023566745197148731
['ClFS1----------'] into ['AAFS1----3A----'], loss: 0.00023673095631175178
['P8YP4---------1'] into ['PLYP4----------'], loss: 0.00023676185077836577
['AAIS2----3A----', 'AAIS2----2A----'] into ['CrIS2----------', 'ClZS2----------'], loss: 0.0002409702870763543
['AUFS1M---------'] into ['NNFS1-----A---1'], loss: 0.00024426151875370575
['PSMP1-P1-------', 'AAMP1----2A----'] into ['AGMP1-----A----', 'AAMP1----3A----'], loss: 0.00024484033094982914
['AAFP2----1N----'] into ['AGFP2-----A----'], loss: 0.0002451343320812255
['AAMP1----1N----'] into ['CrFP1----------', 'AGFP1-----A----'], loss: 0.00024788853929374566
['AAMP2----3A----'] into ['Ca--2----------'], loss: 0.0002542378967566631
['PDFP4----------'] into ['ClHP4----------', 'PKM-4----------'], loss: 0.00025724867473463043
['P1XXXXP3-------'] into ['P1XXXZS3-------'], loss: 0.0002588935045554231
['Dg-------1A---1', 'Dg-------3A---1'] into ['Dg-------1A---3', 'Co-------------'], loss: 0.0002589550998341334
['ClZS7----------'] into ['ClFS7----------'], loss: 0.0002639107342328079
['Cn-P2----------'] into ['CrNP2----------', 'AGNP2-----A----'], loss: 0.0002649749511074971
['PP-P4--2-------'] into ['PP-S4--1-------'], loss: 0.000268647439089649
['P4IS4----------'] into ['P4FP4----------'], loss: 0.00027013929906276194
['PP-P1--1-------'] into ['AAIP1----2A----', 'PP-P1--2-------'], loss: 0.0002763471492169589
['AAMS4----1A----'] into ['P8MS4----------', 'ClMS4----------'], loss: 0.00027644904385682725
['P8FS4----------', 'AAFS4----1N----'] into ['AAFS4----2A----', 'ClFS4----------'], loss: 0.00027714217757024667
['AUNS6M---------', 'CrNS6----------'] into ['PSZS6FS3-------', 'AAIS6----1N----', 'ClZS6----------', 'CrIS6----------', 'PSZS6-P1-------'], loss: 0.0002833672109680205
['AAFS1----2A----'] into ['AAFS1----1N----'], loss: 0.00028385536798410744
['CrNS1----------'] into ['AUNS1M---------'], loss: 0.0002841033468206095
['P8NS4---------1'] into ['AANS4----1N----', 'CrNS4----------'], loss: 0.0002847826421111708
['NNFS4-----N----'] into ['AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1'], loss: 0.00028787490310984735
['NNIS2-----A---1'] into ['ClNS1----------', 'PLNS1---------1'], loss: 0.00028884412054119304
['P8IS4----------'] into ['AAIS4----3A----', 'AGIS4-----A----'], loss: 0.000290818911179244
['P8XP2----------'] into ['PZXP2----------'], loss: 0.0002910759948502109
['PJYS2----------'] into ['Vi-P---2--N----', 'J,-P---1-------'], loss: 0.0002921140232921668
['VpTP---XR-AA--1'] into ['VpNS---XR-AA--1'], loss: 0.00029911558921482274
['PDXP3----------'] into ['ClXP3----------', 'PLXP3----------'], loss: 0.0002997726438863338
['P4FS1----------'] into ['PE--1----------'], loss: 0.0003008320878797009
['PDIP1----------', 'PLIP1----------'] into ['AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------'], loss: 0.000304231086467576
['VB-P---1F-AA---', 'VB-S---1P-NA--1'] into ['VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------'], loss: 0.0003061793502050869
['PPYS1--3-------'] into ['AGIP1-----A----'], loss: 0.00030851448339539226
['AANP7----1A----'] into ['NNIS5-----A----', 'NNFP2-----A---1'], loss: 0.00031213826968988186
['P8FS2---------1'] into ['AAFS2----1N----', 'ClFS2----------'], loss: 0.00031493407391819037
['VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----'] into ['NNFS5-----A----', 'II-------------'], loss: 0.0003174523242206826
['ClZS3----------', 'AGIS2-----A----'] into ['VB-P---2P-NA---', 'PDMS4----------'], loss: 0.00031848083668677206
['P4NS1----------'] into ['P4XP3----------', 'PJXP1----------'], loss: 0.00032714185466686547
['PPFS3--3-------'] into ['PZ--1----------'], loss: 0.00033184553275016483
['CrMS1----------', 'AAMS1----2A----'] into ['AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----'], loss: 0.0003326616795199225
['CrFS1----------'] into ['PSFSXFS3-------'], loss: 0.0003327983565674221
['P9FS6----------', 'P5ZS6--3-------'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----'], loss: 0.00033371520435819535
['PSXP4-P1-------', 'AAFP4----1N----'] into ['PLNS4---------1', 'PZZS7----------', 'PDMP4----------'], loss: 0.0003359470335769679
['PDIS4----------'] into ['CrIS4----------', 'ClIS4----------'], loss: 0.0003401485603056163
['P8FS7---------1'] into ['AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----'], loss: 0.0003419849608371473
['NNNP3-----A----'] into ['NNFS1-----N----', 'NNIP2-----A---1'], loss: 0.0003453450841095695
['PWFP4----------', 'PZFP4----------'] into ['ClHP4----------', 'PKM-4----------', 'PDFP4----------'], loss: 0.0003455456800750222
['ClHP1----------'] into ['PLFP1----------', 'AAFP1----3A----'], loss: 0.0003473558692622784
['NNMS6-----A---1', 'AAMS6----1A----'] into ['NNMP6-----A----', 'NNFP1-----N----'], loss: 0.00034815103931413213
['VB-S---2P-AA---'] into ['AGNS1-----A----'], loss: 0.0003485906053483591
['PSZS2-P1-------'] into ['P8ZS2----------'], loss: 0.0003488404636092586
['PSZS2FS3-------', 'PSZS2-S1-------'] into ['P8ZS2----------', 'PSZS2-P1-------'], loss: 0.00028167265826382253
['VeYS------A----', 'CrIS7----------'] into ['AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----'], loss: 0.0003548361387977095
['AAIS2----1A----'] into ['AUIS2M---------', 'AAIS2----1N----'], loss: 0.0003609439276236123
['P8FS4---------1'] into ['PZFS4----------', 'PWFS4----------', 'PSFS4-P1-------'], loss: 0.00036112443514467766
['AAIS4----1N----', 'PZIS4----------'] into ['AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------'], loss: 0.00036635204344666675
['ClYP4----------'] into ['ClXP4----------'], loss: 0.00036706565100136585
['PLNS4----------', 'AANS1----3A----'] into ['AGNS1-----A----', 'VB-S---2P-AA---'], loss: 0.00036707748228823143
['ClYS1----------'] into ['AGIS1-----A----'], loss: 0.00037002922220758286
['PW--1----------'] into ['PW--4----------'], loss: 0.00037223841660657194
['NNMS3-----A---1'] into ['AAMS3----1A----'], loss: 0.0003723114757928749
['VB-S---3F-NA---'] into ['J^------------1', 'PQ--7----------'], loss: 0.0003724035334773572
['P5ZS3--3-------', 'P5FS3--3-------'] into ['P6-X3----------', 'ACTP------A----', 'PLFS1----------'], loss: 0.00037794169786199347
['Vc-P---1-------'] into ['ACMP------A----'], loss: 0.0003800867254184037
['PZM-1----------'] into ['PP-P3--2-------'], loss: 0.0003881583819167302
['P4YS1----------'] into ['J,-S---1-------', 'P4NS4----------', 'PJYS1----------', 'PJNS1----------'], loss: 0.0003930253142851066
['VB-P---1P-NA---'] into ['VB-S---1P-AA--1'], loss: 0.0003953433978250661
['PPMP1--3-------', 'PZMP1----------'] into ['AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----'], loss: 0.00039909434835052465
['RF-------------'] into ['NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------'], loss: 0.000401458052038866
['P5ZS4--3-------', 'P6-X4----------'] into ['P5XP4--3-------', 'P5FS4--3-------'], loss: 0.0004037776259040245
['VpTP---XR-NA---'] into ['VB-P---3P-NA--1', 'VB-P---3F-NA---'], loss: 0.0004046001536796113
['PPZS7--3-------', 'PPXP2--3-------'] into ['P8ZS3----------', 'PPNS4--3-------'], loss: 0.0004092492513763831
['CrIS1----------', 'AAIS1----3A----'] into ['PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------'], loss: 0.00040970326699290953
['PQ--1----------'] into ['P4XP3----------', 'PJXP1----------', 'P4NS1----------'], loss: 0.0004124207482654181
['A2--------A----', 'AAXXX----1A---8'] into ['NNMXX-----A---8', 'NNIXX-----A---8'], loss: 0.0004149139345738088
['AANS2----1A----'] into ['AUNS2M---------', 'CrNS2----------', 'AANS2----1N----'], loss: 0.00041576859281495376
['NNMS5-----A----'] into ['AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------'], loss: 0.0004246467763391417
['NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1'] into ['CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----'], loss: 0.0004280756591562735
['PDFS1----------'] into ['PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------'], loss: 0.00043474027441732857
['PSYS1-P1-------', 'PSYS1-S1-------'] into ['AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----'], loss: 0.00043499652687142756
['ClXP2----------'] into ['AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------'], loss: 0.00043578802043221614
['PDFS2----------'] into ['PSFS2-P1-------', 'AGFS2-----A----'], loss: 0.0004369551481315506
['PKM-1----------'] into ['PKM-3----------', 'PQ--2----------', 'P4ZS7----------'], loss: 0.0004376604141117077
['Ca--2----------', 'AAMP2----3A----'] into ['PZXP2----------', 'P8XP2----------'], loss: 0.0004428430914211407
['PH-S4--1-------'] into ['Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1'], loss: 0.0004432304414115737
['AAIP2----1A----'] into ['AAIP2----1N----', 'AAIP2----2A----', 'AGIP2-----A----'], loss: 0.0004446777604874691
['AANP6----1A----'] into ['PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------'], loss: 0.00044571297602563273
['Ca--1----------'] into ['Ca--4----------'], loss: 0.0004485041820442622
['PDFS4----------'] into ['AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------'], loss: 0.000451320259733851
['AAIP4----1A----'] into ['PZYP4----------', 'AAIP4----1N----', 'PDIP4----------'], loss: 0.0004605769068917846
['NNMS6-----A----'] into ['NNMP6-----A----', 'NNFP1-----N----', 'NNMS6-----A---1', 'AAMS6----1A----'], loss: 0.00046401985387540005
['PZ--4----------'] into ['ACYS------A----'], loss: 0.00047961092802832096
['Vf--------N----'] into ['AANS1----2A----', 'AANS1----1N----'], loss: 0.00048298850764411224
['P8ZS7----------'] into ['AANP3----1A----'], loss: 0.0004871403553058573
['AAMP7----1A----'] into ['P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----'], loss: 0.0005030576878570095
['PP-P4--1-------'] into ['PP-S4--1-------', 'PP-P4--2-------'], loss: 0.0005042513569848831
['RR--1----------', 'RR--X----------'] into ['NNFS1-----A---1', 'AUFS1M---------'], loss: 0.0005072496345855284
['AGFS1-----A----'] into ['AAFS1----3A----', 'ClFS1----------'], loss: 0.0005091139825422382
['PSHP1-P1-------', 'PZFP1----------'] into ['PLFP1----------', 'AAFP1----3A----', 'ClHP1----------'], loss: 0.0005104549573868638
['VpQW---XR-NA---'] into ['VpQW---XR-AA--1'], loss: 0.0005105955088162892
['PQ--6----------', 'P9XP6----------', 'P9ZS6----------'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----', 'P9FS6----------', 'P5ZS6--3-------'], loss: 0.0005220893249769072
['AAFS4----2A----', 'ClFS4----------', 'P8FS4----------', 'AAFS4----1N----'] into ['AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------', 'PDFS4----------'], loss: 0.0005255893429910495
['PPFS4--3-------'] into ['PHZS4--3-------'], loss: 0.000525953834143153
['ClXP1----------'] into ['AAFP1----2A----', 'AAFP1----1N----'], loss: 0.0005300479197926648
['NNFSX-----A----'] into ['NNIPX-----A----'], loss: 0.000530080612474559
['NNMP2-----A---1', 'PLXP7----------'] into ['AGIP1-----A----', 'PPYS1--3-------'], loss: 0.000531265705439928
['PDMP1----------'] into ['PLMP1----------'], loss: 0.0005319391418313899
['PDXP2----------'] into ['AGFP2-----A----', 'AAFP2----1N----'], loss: 0.0005322017709938525
['PWM-1----------'] into ['PLNS1----------'], loss: 0.000540604911257465
['AAFS7----1A----'] into ['AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1'], loss: 0.0005474023285434483
['PLFP4----------', 'P8FP4---------1'] into ['ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------'], loss: 0.0005480233720124073
['PP-S1--1-------'] into ['AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------'], loss: 0.0005497474311855156
['PPXP4--3-------'] into ['PPZS4--3------2'], loss: 0.0005502797931046697
['VpMP---XR-AA--1'] into ['VpMP---XR-NA---'], loss: 0.0005521782027592512
['PLXP2----------'] into ['PZXP2----------', 'P8XP2----------', 'Ca--2----------', 'AAMP2----3A----'], loss: 0.0005531428213135189
['VB-P---3P-AA--1'] into ['VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1'], loss: 0.0005537818575130785
['AANP4----1A----'] into ['PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----'], loss: 0.0005588926176024263
['AANS4----1A----'] into ['AANS4----1N----', 'CrNS4----------', 'P8NS4---------1'], loss: 0.0005601759243670741
['Vi-P---2--A----', 'Vi-P---1--A----'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----'], loss: 0.000575822707220917
['PSXXXXP3-------'] into ['P1XXXZS3-------', 'P1XXXXP3-------'], loss: 0.000579795243742538
['AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1'] into ['PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------'], loss: 0.000584519090265391
['AAMP2----1A----'] into ['AGMP2-----A----', 'AAMP2----1N----'], loss: 0.0005951677157307982
['NNNP7-----A----'] into ['P6-X7----------', 'PP-P7--1-------'], loss: 0.0005962249134426291
['P5ZS7--3-------', 'P5XP7--3-------'] into ['P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----'], loss: 0.0005897804390823753
['Db------------8'] into ['NNFXX-----A---8'], loss: 0.0005963660423994998
['Dg-------3A----'] into ['ClFS7----------', 'ClZS7----------'], loss: 0.0006083206967914185
['P4MS4----------', 'PJZS3----------', 'P4YP4----------', 'P4FS4----------'] into ['Vi-P---2--N----', 'J,-P---1-------', 'PJYS2----------'], loss: 0.0006106331948039364
['P4FP4----------', 'P4IS4----------'] into ['Vi-P---2--N----', 'J,-P---1-------', 'PJYS2----------', 'P4MS4----------', 'PJZS3----------', 'P4YP4----------', 'P4FS4----------'], loss: 0.0005872867140389662
['AGFS4-----A----'] into ['VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----'], loss: 0.0006199942254642194
['PHZS3--3-------'] into ['PZ--1----------', 'PPFS3--3-------'], loss: 0.0006239881523732995
['Cn-P7----------', 'P8XP7----------', 'ClXP7----------'] into ['P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----'], loss: 0.0006273828521090154
['PPXP3--3-------'] into ['PP-P3--2-------', 'PZM-1----------'], loss: 0.0006296357880983048
['P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1'] into ['P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------'], loss: 0.0006333290990911274
['ClFS6----------', 'PZFS6----------', 'AGIS6-----A----'] into ['P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1'], loss: 0.0006263619565683022
['AAIP6----1A----', 'AAIP6----3A----'] into ['Ca--6----------', 'PZXP6----------', 'PDXP6----------'], loss: 0.0006336693077611706
['P4MP1----------'] into ['P4IP1----------', 'P4FP1----------'], loss: 0.0006380145473267127
['VsMP---XX-AP---'] into ['ACMP------A----', 'Vc-P---1-------'], loss: 0.0006411226378871652
['AAFP6----1A----'] into ['AGFP6-----A----', 'PLXP6----------', 'P8XP6----------'], loss: 0.0006514733867610915
['P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------'] into ['ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1'], loss: 0.0006550263730965035
['ClXP3----------', 'PLXP3----------', 'PDXP3----------'] into ['P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------'], loss: 0.0006592274241572849
['AAIS4----1A----'] into ['AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------'], loss: 0.0006649276231363854
['C}-------------'] into ['NNMXX-----A---8', 'NNIXX-----A---8', 'A2--------A----', 'AAXXX----1A---8'], loss: 0.0006669066563271707
['AAIS1----2A----', 'AAIS1----1N----'] into ['AGIS1-----A----', 'ClYS1----------'], loss: 0.0006787611994773894
['P5XP2--3-------', 'PP-P2--1-------'] into ['ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------'], loss: 0.0006796269759717288
['ClYP1----------'] into ['AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------'], loss: 0.0006796773221613146
['AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------'] into ['PLMP1----------', 'PDMP1----------'], loss: 0.000681926578723245
['PLYS1----------'] into ['PDYS1----------'], loss: 0.0006849175510178647
['PDFP1----------'] into ['PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------'], loss: 0.0006987169761761103
['VB-P---2P-AA---'] into ['VB-S---1P-NA---'], loss: 0.0006999086709006141
['PSFSXFS3-------', 'CrFS1----------'] into ['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----'], loss: 0.000702608746563072
['Vc-S---1-------'] into ['VB-S---1P-AA---'], loss: 0.0007047303753939232
['VpNS---XR-AA--1', 'VpTP---XR-AA--1'] into ['VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---'], loss: 0.0007114893772403401
['PP-P3--1-------'] into ['PH-S3--1-------'], loss: 0.0007121887454635005
['NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----'] into ['J^------------1', 'PQ--7----------', 'VB-S---3F-NA---'], loss: 0.0007145520609939992
['AANS6----1A----'] into ['PSZS6FS3-------', 'AAIS6----1N----', 'ClZS6----------', 'CrIS6----------', 'PSZS6-P1-------', 'AUNS6M---------', 'CrNS6----------'], loss: 0.0007151456072016539
['PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------', 'CrIS1----------', 'AAIS1----3A----'] into ['PDYS1----------', 'PLYS1----------'], loss: 0.0007228456501253088
['PDZS7----------'] into ['AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----', 'VeYS------A----', 'CrIS7----------'], loss: 0.0007235311062447052
['AANS1----1A----'] into ['AUNS1M---------', 'CrNS1----------'], loss: 0.0007324702340073067
['AAMP4----1A----'] into ['PLYP4----------', 'P8YP4---------1'], loss: 0.0007334007106947794
['CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----'] into ['AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----'], loss: 0.0007411117891758807
['Dg-------2A---1'] into ['Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------'], loss: 0.0007675954489430932
['Db------------1'] into ['Cv-------------'], loss: 0.0007724283853681238
['P8XP3----------', 'AAMP3----1A----'] into ['P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------'], loss: 0.0007829677730353791
['NNFPX-----A----'] into ['AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------', 'NNMS5-----A----'], loss: 0.000785682299382591
['ACNS------A----'] into ['AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----'], loss: 0.0007884705329255261
['VB-S---1P-AA--1', 'VB-P---1P-NA---'] into ['VB-S---1P-NA---', 'VB-P---2P-AA---'], loss: 0.0007927509719976223
['CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------'] into ['AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------'], loss: 0.000801198866963955
['AAFS1----1N----', 'AAFS1----2A----'] into ['AANS1----2A----', 'AANS1----1N----', 'Vf--------N----'], loss: 0.0008088697255468915
['Cn-S1----------'] into ['Ca--4----------', 'Ca--1----------'], loss: 0.0008233528556061231
['CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1'] into ['AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------'], loss: 0.0008288299828685312
['PDZS2----------'] into ['P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------'], loss: 0.0008436384949763263
['PE--1----------', 'P4FS1----------'] into ['J,-S---1-------', 'P4NS4----------', 'PJYS1----------', 'PJNS1----------', 'P4YS1----------'], loss: 0.0008461908271523971
['P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------'] into ['PPZS4--3------2', 'PPXP4--3-------'], loss: 0.0008517332081007891
['AAIP3----1A----'] into ['AANP3----1A----', 'P8ZS7----------'], loss: 0.0008816974828067528
['PW--4----------', 'PW--1----------'] into ['PLNS1----------', 'PWM-1----------'], loss: 0.0008826879932773077
['NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------'] into ['AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------'], loss: 0.0008981738855807385
['Dg-------1N----'] into ['ACYS------A----', 'PZ--4----------'], loss: 0.0009049432454271188
['NNMS3-----A----'] into ['AAMS3----1A----', 'NNMS3-----A---1'], loss: 0.000908241324715659
['AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----'] into ['VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----'], loss: 0.00022346334040751177
['VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----', 'AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----'] into ['AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----', 'CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----'], loss: 0.0006149557620880788
['AAMS1----1A----'] into ['AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----', 'PSYS1-P1-------', 'PSYS1-S1-------'], loss: 0.0009091035763646531
['PZFS4----------', 'PWFS4----------', 'PSFS4-P1-------', 'P8FS4---------1'] into ['AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------', 'PDFS4----------', 'AAFS4----2A----', 'ClFS4----------', 'P8FS4----------', 'AAFS4----1N----'], loss: 0.0009131077280775487
['NNIS6-----A----'] into ['NNIS6-----A---1'], loss: 0.0009300286860455947
['NNMP1-----A----'] into ['NNMP1-----A---1'], loss: 0.0009326844847098577
['AAMP1----1A----'] into ['PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------'], loss: 0.0009270809523082175
['PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----'] into ['AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----'], loss: 0.0009471533147338215
['NNFP6-----A----'] into ['NNNP6-----A----'], loss: 0.0008076195170246304
['ClXP4----------', 'ClYP4----------'] into ['PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----'], loss: 0.0009677592271756449
['PQ--4----------'] into ['PKM-3----------', 'PQ--2----------', 'P4ZS7----------', 'PKM-1----------'], loss: 0.0009678295910635392
['RV--3----------'] into ['RR--3----------'], loss: 0.0009720246510990481
['VpYS---XR-AA--1'] into ['VpYS---XR-AA---'], loss: 0.0009720827141094402
['AAFP3----1A----'] into ['NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----'], loss: 0.0009863086792256063
['AGFP2-----A----', 'AAFP2----1N----', 'PDXP2----------'] into ['AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------', 'CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------'], loss: 0.001028480803557828
['VB-P---3P-NA---'] into ['VpMP---XR-NA---', 'VpMP---XR-AA--1'], loss: 0.0010347891979745867
['VpTP---XR-AA---'] into ['VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1'], loss: 0.0010454767695581492
['ClXS2----------'] into ['PZXP2----------', 'P8XP2----------', 'Ca--2----------', 'AAMP2----3A----', 'PLXP2----------'], loss: 0.0010584317386930983
['Cn-S4----------'] into ['Ca--4----------', 'Ca--1----------', 'Cn-S1----------'], loss: 0.0010572540924380024
['PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------'] into ['PH-S3--1-------', 'PP-P3--1-------'], loss: 0.0010606584189547799
['NNIPX-----A----', 'NNFSX-----A----'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----', 'Vi-P---2--A----', 'Vi-P---1--A----'], loss: 0.0010613189461292204
['PHZS4--3-------', 'PPFS4--3-------'] into ['PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------'], loss: 0.001071331787781754
['VB-P---3F-AA---'] into ['VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1'], loss: 0.0010715906786546224
['NNFS1-----A---1', 'AUFS1M---------', 'RR--1----------', 'RR--X----------'] into ['AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------', 'NNMS5-----A----', 'NNFPX-----A----'], loss: 0.0010835893698258648
['Vi-P---2--N----', 'J,-P---1-------', 'PJYS2----------', 'P4MS4----------', 'PJZS3----------', 'P4YP4----------', 'P4FS4----------', 'P4FP4----------', 'P4IS4----------'] into ['PKM-3----------', 'PQ--2----------', 'P4ZS7----------', 'PKM-1----------', 'PQ--4----------'], loss: 0.0010925690102154889
['RR--4----------'] into ['RV--4----------'], loss: 0.001091856102494451
['AAFP4----1A----'] into ['ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1'], loss: 0.0010787855607888148
['PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------'] into ['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------'], loss: 0.0010948987048702687
['VpNS---XR-NA---'] into ['J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----'], loss: 0.0010976803197157842
['NNFXX-----A---8', 'Db------------8'] into ['NNMXX-----A---8', 'NNIXX-----A---8', 'A2--------A----', 'AAXXX----1A---8', 'C}-------------'], loss: 0.0010991340184958454
['PDZS6----------'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----', 'P9FS6----------', 'P5ZS6--3-------', 'PQ--6----------', 'P9XP6----------', 'P9ZS6----------'], loss: 0.0011118594208968647
['NNNP4-----A----'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------'], loss: 0.0011222186708015267
['NNMP7-----A----'] into ['P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----', 'P5ZS7--3-------', 'P5XP7--3-------'], loss: 0.0011391617719325843
['PDZS3----------'] into ['AAIS3----1A----'], loss: 0.0011478163950671295
['NNMS4-----A----'] into ['P8MS4----------', 'ClMS4----------', 'AAMS4----1A----'], loss: 0.0011495274990832938
['Dg-------2A----'] into ['Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1'], loss: 0.001208213008683829
['PDNS4----------'] into ['PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------'], loss: 0.0012135337520150671
['AAFP7----2A----', 'AAFP7----1A----'] into ['P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----', 'Cn-P7----------', 'P8XP7----------', 'ClXP7----------'], loss: 0.0012395555713032406
['P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------'] into ['P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------'], loss: 0.0012437291554775647
['RV--2----------'] into ['AAFS2----2A----', 'CrFS2----------', 'AUFS2M---------'], loss: 0.0012910226430475071
['J,-X---3-------'] into ['P4XP3----------', 'PJXP1----------', 'P4NS1----------', 'PQ--1----------'], loss: 0.001297213133604938
['AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------'] into ['AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------'], loss: 0.0013082914260193125
['NNFP3-----A----'] into ['NNIP3-----A----'], loss: 0.0013278617870135512
['NNIP6-----A----'] into ['NNMP6-----A----', 'NNFP1-----N----', 'NNMS6-----A---1', 'AAMS6----1A----', 'NNMS6-----A----'], loss: 0.001331602242482039
['ClFS7----------', 'ClZS7----------', 'Dg-------3A----'] into ['AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----'], loss: 0.0013331746744274784
['NNFP7-----A----'] into ['P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----', 'P5ZS7--3-------', 'P5XP7--3-------', 'NNMP7-----A----'], loss: 0.0013340080057473464
['AAFS4----1A----'] into ['AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------', 'PDFS4----------', 'AAFS4----2A----', 'ClFS4----------', 'P8FS4----------', 'AAFS4----1N----', 'PZFS4----------', 'PWFS4----------', 'PSFS4-P1-------', 'P8FS4---------1'], loss: 0.0013512390984049827
['VsYS---XX-AP---'] into ['AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----'], loss: 0.0013716335403436957
['PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------'] into ['AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1'], loss: 0.0014110146924557147
['AAIP1----1A----'] into ['AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------'], loss: 0.001411294796742799
['RV--7----------'] into ['RR--7----------'], loss: 0.0014301735744251046
['P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----'] into ['AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----'], loss: 0.0014267693093746618
['NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----'] into ['AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----'], loss: 0.0013973237713253584
['NNMP3-----A----'] into ['NNIP3-----A----', 'NNFP3-----A----'], loss: 0.0014376503637783472
['VB-S---3F-AA---'] into ['VpQW---XR-AA--1', 'VpQW---XR-NA---'], loss: 0.001443511721998469
['VB-S---3P-NA---'] into ['VpYS---XR-NA---'], loss: 0.001466076376956317
['ACYS------A----', 'PZ--4----------', 'Dg-------1N----'] into ['Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----'], loss: 0.0014756253365247167
['Db-------------'] into ['Cv-------------', 'Db------------1'], loss: 0.0014568770490205889
['PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------'] into ['PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------'], loss: 0.001491775704238096
['PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------'] into ['PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------'], loss: 0.0014091573140008855
['VsTP---XX-AP---'] into ['ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---'], loss: 0.0014869651192212121
['AANS7----1N----', 'AANS7----1A----'] into ['AAMS7----1A----', 'AAMS7----3A----'], loss: 0.0014922318651327373
['XX-------------'] into ['AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------', 'NNMS5-----A----', 'NNFPX-----A----', 'NNFS1-----A---1', 'AUFS1M---------', 'RR--1----------', 'RR--X----------'], loss: 0.001512015671686637
['PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----'], loss: 0.0015503766251371027
['VB-P---1P-AA---'] into ['VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---'], loss: 0.0015628283859096049
['VpMP---XR-AA---'] into ['VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---'], loss: 0.0015568230038479204
['AAIS7----1A----'] into ['AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----', 'VeYS------A----', 'CrIS7----------', 'PDZS7----------'], loss: 0.001589652404592011
['VsQW---XX-AP---'] into ['VsNS---XX-AP---'], loss: 0.0015916369539163988
['PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----'] into ['PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------'], loss: 0.0016087858883503184
['NNIP4-----A----'] into ['NNMP4-----A----'], loss: 0.0012874379529257789
['NNNXX-----A----'] into ['NNIXX-----A----'], loss: 0.0016483399476322279
['PDFS6----------', 'AAFS6----3A----'] into ['Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----'], loss: 0.0016551534144433732
['NNNS3-----A----'] into ['AANS3----1A----'], loss: 0.0016702177818335457
['AAIP7----1A----'] into ['P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----', 'Cn-P7----------', 'P8XP7----------', 'ClXP7----------', 'AAFP7----2A----', 'AAFP7----1A----'], loss: 0.0016859543126698363
['P4IP1----------', 'P4FP1----------', 'P4MP1----------'] into ['P4XP3----------', 'PJXP1----------', 'P4NS1----------', 'PQ--1----------', 'J,-X---3-------'], loss: 0.0017225721488248258
['AAXXX----1A----'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----', 'Vi-P---2--A----', 'Vi-P---1--A----', 'NNIPX-----A----', 'NNFSX-----A----'], loss: 0.0017406687203390218
['AANS1----2A----', 'AANS1----1N----', 'Vf--------N----', 'AAFS1----1N----', 'AAFS1----2A----'] into ['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------', 'PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------'], loss: 0.0017446691137504942
['AANP1----1A----', 'CrNP1----------'] into ['CrIS4----------', 'ClIS4----------', 'PDIS4----------'], loss: 0.0017408085304985686
['VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---'] into ['VpYS---XR-NA---', 'VB-S---3P-NA---'], loss: 0.0018758592295739182
['VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---'] into ['J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---'], loss: 0.0019146421343032968
['NNMS2-----A----'] into ['AAMS2----1A----'], loss: 0.001957230268844726
['AAMS2----1A----', 'NNMS2-----A----'] into ['PDFS3----------', 'AAFS3----1A----'], loss: -0.00804188814726723
['PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----'] into ['P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------'], loss: -0.010406776425865733
['P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----'] into ['P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----'], loss: -0.008911437946749276
['P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----'] into ['Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----'], loss: -0.01026459046954874
['Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----'] into ['AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----'], loss: -0.007629447183640893
['AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----', 'Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----', 'PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----'], loss: 5.7883248249381014e-05
['NNNP6-----A----', 'NNFP6-----A----'] into ['NNMP6-----A----', 'NNFP1-----N----', 'NNMS6-----A---1', 'AAMS6----1A----', 'NNMS6-----A----', 'NNIP6-----A----'], loss: 0.0011663564248341196
['RV--6----------'] into ['RR--6----------'], loss: 0.0019047440579213247
['RR--2----------'] into ['AAFS2----2A----', 'CrFS2----------', 'AUFS2M---------', 'RV--2----------'], loss: 0.0019799047105345693
['ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------'] into ['PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------'], loss: 0.0020028443367202326
['J,-S---1-------', 'P4NS4----------', 'PJYS1----------', 'PJNS1----------', 'P4YS1----------', 'PE--1----------', 'P4FS1----------'] into ['PKM-3----------', 'PQ--2----------', 'P4ZS7----------', 'PKM-1----------', 'PQ--4----------', 'Vi-P---2--N----', 'J,-P---1-------', 'PJYS2----------', 'P4MS4----------', 'PJZS3----------', 'P4YP4----------', 'P4FS4----------', 'P4FP4----------', 'P4IS4----------'], loss: 0.00200313036417176
['AAFP1----1A----'] into ['AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------'], loss: 0.0021046427834323436
['Xx-------------'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----', 'Vi-P---2--A----', 'Vi-P---1--A----', 'NNIPX-----A----', 'NNFSX-----A----', 'AAXXX----1A----'], loss: 0.0021111332534291594
['AAIS1----1A----'] into ['PDYS1----------', 'PLYS1----------', 'PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------', 'CrIS1----------', 'AAIS1----3A----'], loss: 0.0021225545558653655
['TT-------------'] into ['AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----'], loss: 0.002138270823056438
['AANP2----1A----'] into ['AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------', 'CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------', 'AGFP2-----A----', 'AAFP2----1N----', 'PDXP2----------'], loss: 0.002144593858427937
['PZXP2----------', 'P8XP2----------', 'Ca--2----------', 'AAMP2----3A----', 'PLXP2----------', 'ClXS2----------'] into ['AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------', 'CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------', 'AGFP2-----A----', 'AAFP2----1N----', 'PDXP2----------', 'AANP2----1A----'], loss: 0.0016540799332079182
['X@-------------'] into ['AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------', 'NNMS5-----A----', 'NNFPX-----A----', 'NNFS1-----A---1', 'AUFS1M---------', 'RR--1----------', 'RR--X----------', 'XX-------------'], loss: 0.002221545064182193
['VB-P---3P-AA---'] into ['VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---'], loss: 0.002363733209273179
['NNFS3-----A----'] into ['NNIP3-----A----', 'NNFP3-----A----', 'NNMP3-----A----'], loss: 0.0023918863503050156
['Dg-------1A----'] into ['Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----'], loss: 0.0024218556088530393
['AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------', 'AAIS4----1A----'] into ['CrIS4----------', 'ClIS4----------', 'PDIS4----------', 'AANP1----1A----', 'CrNP1----------'], loss: 0.002434361203902464
['AAFS1----1A----'] into ['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------', 'PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------', 'AANS1----2A----', 'AANS1----1N----', 'Vf--------N----', 'AAFS1----1N----', 'AAFS1----2A----'], loss: 0.0025174246017960283
['P7-X4----------'] into ['P7-X3----------'], loss: 0.002539340066858116
['NNNS7-----A----'] into ['NNMS7-----A----'], loss: 0.0025909236978939884
['NNNP1-----A----'] into ['NNFP1-----A----'], loss: 0.0026402216063716073
['AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----', 'CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----', 'VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----', 'AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----'] into ['NNIP7-----A----'], loss: 0.002651228900769199
['AAIS3----1A----', 'PDZS3----------'] into ['AANS3----1A----', 'NNNS3-----A----'], loss: 0.0026941076748845934
['P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----'] into ['AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----'], loss: 0.0027814373308090844
['AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---'] into ['VsNS---XX-AP---', 'VsQW---XX-AP---'], loss: 0.0028207486544391215
['NNFXX-----A----'] into ['NNIXX-----A----', 'NNNXX-----A----'], loss: 0.0028517832737261764
['VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---'] into ['J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---'], loss: 0.002869407565843583
['PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------'] into ['PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------'], loss: 0.0028738946647087924
['AAIS6----1A----'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----', 'P9FS6----------', 'P5ZS6--3-------', 'PQ--6----------', 'P9XP6----------', 'P9ZS6----------', 'PDZS6----------'], loss: 0.0028910280042056136
['NNIS3-----A----'] into ['AANS3----1A----', 'NNNS3-----A----', 'AAIS3----1A----', 'PDZS3----------'], loss: 0.002920633954280995
['NNNP2-----A----'] into ['NNMP2-----A----'], loss: 0.00292776059253616
['AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------'] into ['PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------'], loss: 0.0029433270588885138
['Ca--4----------', 'Ca--1----------', 'Cn-S1----------', 'Cn-S4----------'] into ['AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------', 'CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------', 'AGFP2-----A----', 'AAFP2----1N----', 'PDXP2----------', 'AANP2----1A----', 'PZXP2----------', 'P8XP2----------', 'Ca--2----------', 'AAMP2----3A----', 'PLXP2----------', 'ClXS2----------'], loss: 0.002996828570055904
['Cv-------------', 'Db------------1', 'Db-------------'] into ['AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------'], loss: 0.003084239923887796
['Vf--------A----'] into ['ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---'], loss: 0.003177510923646168
['VpNS---XR-AA---'] into ['VpQW---XR-AA---'], loss: 0.0033277573775163153
['PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1'] into ['P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------'], loss: 0.003349431266080769
['P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----', 'P5ZS7--3-------', 'P5XP7--3-------', 'NNMP7-----A----', 'NNFP7-----A----'] into ['NNIP7-----A----', 'AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----', 'CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----', 'VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----', 'AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----'], loss: 0.0034034502774638645
['Vc-X---3-------'] into ['VB-S---1P-AA---', 'Vc-S---1-------'], loss: 0.0036265218164639795
['PDNS1----------'] into ['PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------'], loss: 0.0036396589673149166
['NNFP4-----A----'] into ['NNMP4-----A----', 'NNIP4-----A----'], loss: 0.003708267388660934
['AAMS7----1A----', 'AAMS7----3A----', 'AANS7----1N----', 'AANS7----1A----'] into ['P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----', 'Cn-P7----------', 'P8XP7----------', 'ClXP7----------', 'AAFP7----2A----', 'AAFP7----1A----', 'AAIP7----1A----'], loss: 0.003772852648616518
['NNMS7-----A----', 'NNNS7-----A----'] into ['NNIP7-----A----', 'AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----', 'CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----', 'VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----', 'AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----', 'P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----', 'P5ZS7--3-------', 'P5XP7--3-------', 'NNMP7-----A----', 'NNFP7-----A----'], loss: 0.00345299100323309
['NNMXX-----A---8', 'NNIXX-----A---8', 'A2--------A----', 'AAXXX----1A---8', 'C}-------------', 'NNFXX-----A---8', 'Db------------8'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----', 'Vi-P---2--A----', 'Vi-P---1--A----', 'NNIPX-----A----', 'NNFSX-----A----', 'AAXXX----1A----', 'Xx-------------'], loss: 0.003860495333398277
['AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------', 'NNMS5-----A----', 'NNFPX-----A----', 'NNFS1-----A---1', 'AUFS1M---------', 'RR--1----------', 'RR--X----------', 'XX-------------', 'X@-------------'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----', 'Vi-P---2--A----', 'Vi-P---1--A----', 'NNIPX-----A----', 'NNFSX-----A----', 'AAXXX----1A----', 'Xx-------------', 'NNMXX-----A---8', 'NNIXX-----A---8', 'A2--------A----', 'AAXXX----1A---8', 'C}-------------', 'NNFXX-----A---8', 'Db------------8'], loss: 0.0031507994832887555
['VB-S---3P-AA---'] into ['J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---'], loss: 0.0038025972029711867
['VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---', 'VB-P---1P-AA---'] into ['VB-S---1P-AA---', 'Vc-S---1-------', 'Vc-X---3-------'], loss: 0.0038326308515695046
['PSXXXZS3-------'] into ['P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1'], loss: 0.0038632069490340654
['PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----'] into ['AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----'], loss: 0.0039062044126886917
['J,-------------'] into ['P4XP3----------', 'PJXP1----------', 'P4NS1----------', 'PQ--1----------', 'J,-X---3-------', 'P4IP1----------', 'P4FP1----------', 'P4MP1----------'], loss: 0.003944845783473494
['NNNS4-----A----'] into ['AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----'], loss: 0.00402806841086859
['PKM-3----------', 'PQ--2----------', 'P4ZS7----------', 'PKM-1----------', 'PQ--4----------', 'Vi-P---2--N----', 'J,-P---1-------', 'PJYS2----------', 'P4MS4----------', 'PJZS3----------', 'P4YP4----------', 'P4FS4----------', 'P4FP4----------', 'P4IS4----------', 'J,-S---1-------', 'P4NS4----------', 'PJYS1----------', 'PJNS1----------', 'P4YS1----------', 'PE--1----------', 'P4FS1----------'] into ['P4XP3----------', 'PJXP1----------', 'P4NS1----------', 'PQ--1----------', 'J,-X---3-------', 'P4IP1----------', 'P4FP1----------', 'P4MP1----------', 'J,-------------'], loss: 0.004128475598082032
['AAIP2----1N----', 'AAIP2----2A----', 'AGIP2-----A----', 'AAIP2----1A----'] into ['AGMP2-----A----', 'AAMP2----1N----', 'AAMP2----1A----'], loss: 0.004167584994181479
['NNIP2-----A----'] into ['NNMP2-----A----', 'NNNP2-----A----'], loss: 0.0022591868825844463
['AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------', 'CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------', 'AGFP2-----A----', 'AAFP2----1N----', 'PDXP2----------', 'AANP2----1A----', 'PZXP2----------', 'P8XP2----------', 'Ca--2----------', 'AAMP2----3A----', 'PLXP2----------', 'ClXS2----------', 'Ca--4----------', 'Ca--1----------', 'Cn-S1----------', 'Cn-S4----------'] into ['AGMP2-----A----', 'AAMP2----1N----', 'AAMP2----1A----', 'AAIP2----1N----', 'AAIP2----2A----', 'AGIP2-----A----', 'AAIP2----1A----'], loss: 0.003721809233251873
['NNNS6-----A----'] into ['NNIS6-----A---1', 'NNIS6-----A----'], loss: 0.004142942700799372
['PSZS6FS3-------', 'AAIS6----1N----', 'ClZS6----------', 'CrIS6----------', 'PSZS6-P1-------', 'AUNS6M---------', 'CrNS6----------', 'AANS6----1A----'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----', 'P9FS6----------', 'P5ZS6--3-------', 'PQ--6----------', 'P9XP6----------', 'P9ZS6----------', 'PDZS6----------', 'AAIS6----1A----'], loss: 0.0014645681962276657
['Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----'] into ['AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------'], loss: 0.004134441800701419
['NNFS7-----A----'] into ['NNIS7-----A----'], loss: 0.004364474442160471
['NNMS1-----A----'] into ['AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----', 'PSYS1-P1-------', 'PSYS1-S1-------', 'AAMS1----1A----'], loss: 0.004586127437324053
['AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----', 'PSYS1-P1-------', 'PSYS1-S1-------', 'AAMS1----1A----', 'NNMS1-----A----'] into ['AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----', 'VeYS------A----', 'CrIS7----------', 'PDZS7----------', 'AAIS7----1A----'], loss: -0.0039883901140692415
['AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----', 'VeYS------A----', 'CrIS7----------', 'PDZS7----------', 'AAIS7----1A----', 'AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----', 'PSYS1-P1-------', 'PSYS1-S1-------', 'AAMS1----1A----', 'NNMS1-----A----'] into ['P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----', 'Cn-P7----------', 'P8XP7----------', 'ClXP7----------', 'AAFP7----2A----', 'AAFP7----1A----', 'AAIP7----1A----', 'AAMS7----1A----', 'AAMS7----3A----', 'AANS7----1N----', 'AANS7----1A----'], loss: -0.011550171667215305
['P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----', 'Cn-P7----------', 'P8XP7----------', 'ClXP7----------', 'AAFP7----2A----', 'AAFP7----1A----', 'AAIP7----1A----', 'AAMS7----1A----', 'AAMS7----3A----', 'AANS7----1N----', 'AANS7----1A----', 'AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----', 'VeYS------A----', 'CrIS7----------', 'PDZS7----------', 'AAIS7----1A----', 'AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----', 'PSYS1-P1-------', 'PSYS1-S1-------', 'AAMS1----1A----', 'NNMS1-----A----'] into ['NNIP7-----A----', 'AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----', 'CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----', 'VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----', 'AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----', 'P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----', 'P5ZS7--3-------', 'P5XP7--3-------', 'NNMP7-----A----', 'NNFP7-----A----', 'NNMS7-----A----', 'NNNS7-----A----'], loss: -0.008934883084231324
['NNIXX-----A----', 'NNNXX-----A----', 'NNFXX-----A----'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----', 'Vi-P---2--A----', 'Vi-P---1--A----', 'NNIPX-----A----', 'NNFSX-----A----', 'AAXXX----1A----', 'Xx-------------', 'NNMXX-----A---8', 'NNIXX-----A---8', 'A2--------A----', 'AAXXX----1A---8', 'C}-------------', 'NNFXX-----A---8', 'Db------------8', 'AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------', 'NNMS5-----A----', 'NNFPX-----A----', 'NNFS1-----A---1', 'AUFS1M---------', 'RR--1----------', 'RR--X----------', 'XX-------------', 'X@-------------'], loss: 0.0046303039209365515
['AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----'] into ['AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----'], loss: 0.004656173386433508
['NNIP3-----A----', 'NNFP3-----A----', 'NNMP3-----A----', 'NNFS3-----A----'] into ['AANS3----1A----', 'NNNS3-----A----', 'AAIS3----1A----', 'PDZS3----------', 'NNIS3-----A----'], loss: 0.0047029881575612765
['PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----'] into ['ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----'], loss: 0.004877333086881201
['PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------'] into ['PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------'], loss: 0.00515065552577383
['VpYS---XR-AA---', 'VpYS---XR-AA--1'] into ['VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---'], loss: 0.005399945511679554
['NNIP1-----A----'] into ['AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----'], loss: 0.005814647418408689
['AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----'] into ['AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----'], loss: 0.00213146516251325
['AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----'] into ['P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------'], loss: 0.005968599684499783
['VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---'] into ['VpQW---XR-AA---', 'VpNS---XR-AA---'], loss: 0.006077094996580996
['AUNS1M---------', 'CrNS1----------', 'AANS1----1A----'] into ['VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---'], loss: 0.006276290998991909
['NNNS1-----A----'] into ['NNMP1-----A---1', 'NNMP1-----A----'], loss: 0.006494455113920047
['J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---', 'VB-S---3P-AA---'] into ['VpQW---XR-AA---', 'VpNS---XR-AA---', 'VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---'], loss: 0.006237120943834685
['NNIS2-----A----'] into ['NNNS2-----A----'], loss: 0.006493621379351295
['NNFP2-----A----'] into ['NNMP2-----A----', 'NNNP2-----A----', 'NNIP2-----A----'], loss: 0.0066410293766830195
['AAFP2----1A----'] into ['AGMP2-----A----', 'AAMP2----1N----', 'AAMP2----1A----', 'AAIP2----1N----', 'AAIP2----2A----', 'AGIP2-----A----', 'AAIP2----1A----', 'AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------', 'CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------', 'AGFP2-----A----', 'AAFP2----1N----', 'PDXP2----------', 'AANP2----1A----', 'PZXP2----------', 'P8XP2----------', 'Ca--2----------', 'AAMP2----3A----', 'PLXP2----------', 'ClXS2----------', 'Ca--4----------', 'Ca--1----------', 'Cn-S1----------', 'Cn-S4----------'], loss: 0.0013400990645433502
['AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------', 'Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----'] into ['PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------', 'PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------'], loss: 0.006654161283370457
['NNIS4-----A----'] into ['AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----', 'NNNS4-----A----'], loss: 0.00702189846250178
['CrIS4----------', 'ClIS4----------', 'PDIS4----------', 'AANP1----1A----', 'CrNP1----------', 'AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------', 'AAIS4----1A----'] into ['AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----', 'NNNS4-----A----', 'NNIS4-----A----'], loss: 0.00799569906598508
['ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----'] into ['VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----'], loss: 0.007977778186216422
['NNMP6-----A----', 'NNFP1-----N----', 'NNMS6-----A---1', 'AAMS6----1A----', 'NNMS6-----A----', 'NNIP6-----A----', 'NNNP6-----A----', 'NNFP6-----A----'] into ['NNIS6-----A---1', 'NNIS6-----A----', 'NNNS6-----A----'], loss: 0.00801353870834221
['VB-S---1P-AA---', 'Vc-S---1-------', 'Vc-X---3-------', 'VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---', 'VB-P---1P-AA---'] into ['PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------', 'PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------', 'AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------', 'Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----'], loss: 0.008202285323553793
['NNFP1-----A----', 'NNNP1-----A----'] into ['NNMP1-----A---1', 'NNMP1-----A----', 'NNNS1-----A----'], loss: 0.008338979431132945
['AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------', 'AAFP1----1A----'] into ['AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----', 'AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----'], loss: 0.00724492951290308
['NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----'] into ['ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----'], loss: 0.009695361773092732
['ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----'] into ['P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----'], loss: -0.00026655805131180557
['P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----'] into ['NNNS2-----A----', 'NNIS2-----A----'], loss: -0.005252071569042942
['NNNS2-----A----', 'NNIS2-----A----', 'P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----'] into ['VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----', 'ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----'], loss: 0.009296354263374159
['VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----', 'ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----', 'NNNS2-----A----', 'NNIS2-----A----', 'P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----'] into ['AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----', 'AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----', 'AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------', 'AAFP1----1A----'], loss: 0.00992904682477062
['NNIS7-----A----', 'NNFS7-----A----'] into ['NNIP7-----A----', 'AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----', 'CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----', 'VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----', 'AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----', 'P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----', 'P5ZS7--3-------', 'P5XP7--3-------', 'NNMP7-----A----', 'NNFP7-----A----', 'NNMS7-----A----', 'NNNS7-----A----', 'P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----', 'Cn-P7----------', 'P8XP7----------', 'ClXP7----------', 'AAFP7----2A----', 'AAFP7----1A----', 'AAIP7----1A----', 'AAMS7----1A----', 'AAMS7----3A----', 'AANS7----1N----', 'AANS7----1A----', 'AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----', 'VeYS------A----', 'CrIS7----------', 'PDZS7----------', 'AAIS7----1A----', 'AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----', 'PSYS1-P1-------', 'PSYS1-S1-------', 'AAMS1----1A----', 'NNMS1-----A----'], loss: 0.009486509218816793
['PSFS6-P1-------', 'AAFS6----1N----', 'CrFS6----------', 'AUFS6M---------', 'AAFS6----1A----'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----', 'P9FS6----------', 'P5ZS6--3-------', 'PQ--6----------', 'P9XP6----------', 'P9ZS6----------', 'PDZS6----------', 'AAIS6----1A----', 'PSZS6FS3-------', 'AAIS6----1N----', 'ClZS6----------', 'CrIS6----------', 'PSZS6-P1-------', 'AUNS6M---------', 'CrNS6----------', 'AANS6----1A----'], loss: 0.01009637654691997
['RR--3----------', 'RV--3----------'] into ['RR--7----------', 'RV--7----------'], loss: 0.010563693497178698
['AANS3----1A----', 'NNNS3-----A----', 'AAIS3----1A----', 'PDZS3----------', 'NNIS3-----A----', 'NNIP3-----A----', 'NNFP3-----A----', 'NNMP3-----A----', 'NNFS3-----A----'] into ['NNIP7-----A----', 'AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----', 'CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----', 'VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----', 'AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----', 'P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----', 'P5ZS7--3-------', 'P5XP7--3-------', 'NNMP7-----A----', 'NNFP7-----A----', 'NNMS7-----A----', 'NNNS7-----A----', 'P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----', 'Cn-P7----------', 'P8XP7----------', 'ClXP7----------', 'AAFP7----2A----', 'AAFP7----1A----', 'AAIP7----1A----', 'AAMS7----1A----', 'AAMS7----3A----', 'AANS7----1N----', 'AANS7----1A----', 'AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----', 'VeYS------A----', 'CrIS7----------', 'PDZS7----------', 'AAIS7----1A----', 'AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----', 'PSYS1-P1-------', 'PSYS1-S1-------', 'AAMS1----1A----', 'NNMS1-----A----', 'NNIS7-----A----', 'NNFS7-----A----'], loss: 0.010315021337354069
['NNFS1-----A----'] into ['NNIS1-----A----'], loss: 0.011958125126148404
['PDYS1----------', 'PLYS1----------', 'PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------', 'CrIS1----------', 'AAIS1----3A----', 'AAIS1----1A----'] into ['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------', 'PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------', 'AANS1----2A----', 'AANS1----1N----', 'Vf--------N----', 'AAFS1----1N----', 'AAFS1----2A----', 'AAFS1----1A----'], loss: 0.003227865282079747
['C=-------------'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----', 'Vi-P---2--A----', 'Vi-P---1--A----', 'NNIPX-----A----', 'NNFSX-----A----', 'AAXXX----1A----', 'Xx-------------', 'NNMXX-----A---8', 'NNIXX-----A---8', 'A2--------A----', 'AAXXX----1A---8', 'C}-------------', 'NNFXX-----A---8', 'Db------------8', 'AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------', 'NNMS5-----A----', 'NNFPX-----A----', 'NNFS1-----A---1', 'AUFS1M---------', 'RR--1----------', 'RR--X----------', 'XX-------------', 'X@-------------', 'NNIXX-----A----', 'NNNXX-----A----', 'NNFXX-----A----'], loss: 0.011764928005260918
['RV--4----------', 'RR--4----------'] into ['AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------', 'PDFS4----------', 'AAFS4----2A----', 'ClFS4----------', 'P8FS4----------', 'AAFS4----1N----', 'PZFS4----------', 'PWFS4----------', 'PSFS4-P1-------', 'P8FS4---------1', 'AAFS4----1A----'], loss: 0.013272332397128817
['NNMP1-----A---1', 'NNMP1-----A----', 'NNNS1-----A----', 'NNFP1-----A----', 'NNNP1-----A----'] into ['NNIS1-----A----', 'NNFS1-----A----'], loss: 0.014548344718174013
['NNFS2-----A----'] into ['AUIS2M---------', 'AAIS2----1N----', 'AAIS2----1A----'], loss: 0.015582456763485072
['J^-------------'] into ['P7-X3----------', 'P7-X4----------'], loss: 0.016350975646489033
['VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---', 'VpYS---XR-AA---', 'VpYS---XR-AA--1'] into ['P7-X3----------', 'P7-X4----------', 'J^-------------'], loss: 0.0161603278774853
['VpQW---XR-AA---', 'VpNS---XR-AA---', 'VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---', 'J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---', 'VB-S---3P-AA---'] into ['P7-X3----------', 'P7-X4----------', 'J^-------------', 'VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---', 'VpYS---XR-AA---', 'VpYS---XR-AA--1'], loss: 0.01182389683268497
['NNFS6-----A----'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----', 'P9FS6----------', 'P5ZS6--3-------', 'PQ--6----------', 'P9XP6----------', 'P9ZS6----------', 'PDZS6----------', 'AAIS6----1A----', 'PSZS6FS3-------', 'AAIS6----1N----', 'ClZS6----------', 'CrIS6----------', 'PSZS6-P1-------', 'AUNS6M---------', 'CrNS6----------', 'AANS6----1A----', 'PSFS6-P1-------', 'AAFS6----1N----', 'CrFS6----------', 'AUFS6M---------', 'AAFS6----1A----'], loss: 0.016702297363430107
['NNIS6-----A---1', 'NNIS6-----A----', 'NNNS6-----A----', 'NNMP6-----A----', 'NNFP1-----N----', 'NNMS6-----A---1', 'AAMS6----1A----', 'NNMS6-----A----', 'NNIP6-----A----', 'NNNP6-----A----', 'NNFP6-----A----'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----', 'P9FS6----------', 'P5ZS6--3-------', 'PQ--6----------', 'P9XP6----------', 'P9ZS6----------', 'PDZS6----------', 'AAIS6----1A----', 'PSZS6FS3-------', 'AAIS6----1N----', 'ClZS6----------', 'CrIS6----------', 'PSZS6-P1-------', 'AUNS6M---------', 'CrNS6----------', 'AANS6----1A----', 'PSFS6-P1-------', 'AAFS6----1N----', 'CrFS6----------', 'AUFS6M---------', 'AAFS6----1A----', 'NNFS6-----A----'], loss: 0.010917748865734248
['NNFS4-----A----'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----', 'PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----', 'AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----', 'Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----'], loss: 0.019925709511248012
['AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----', 'NNNS4-----A----', 'NNIS4-----A----', 'CrIS4----------', 'ClIS4----------', 'PDIS4----------', 'AANP1----1A----', 'CrNP1----------', 'AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------', 'AAIS4----1A----'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----', 'PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----', 'AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----', 'Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----', 'NNFS4-----A----'], loss: 0.015464270984066511
['PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------', 'PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------', 'AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------', 'Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----', 'VB-S---1P-AA---', 'Vc-S---1-------', 'Vc-X---3-------', 'VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---', 'VB-P---1P-AA---'] into ['P7-X3----------', 'P7-X4----------', 'J^-------------', 'VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---', 'VpYS---XR-AA---', 'VpYS---XR-AA--1', 'VpQW---XR-AA---', 'VpNS---XR-AA---', 'VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---', 'J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---', 'VB-S---3P-AA---'], loss: 0.022196162305820416
['P7-X3----------', 'P7-X4----------', 'J^-------------', 'VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---', 'VpYS---XR-AA---', 'VpYS---XR-AA--1', 'VpQW---XR-AA---', 'VpNS---XR-AA---', 'VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---', 'J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---', 'VB-S---3P-AA---', 'PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------', 'PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------', 'AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------', 'Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----', 'VB-S---1P-AA---', 'Vc-S---1-------', 'Vc-X---3-------', 'VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---', 'VB-P---1P-AA---'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----', 'PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----', 'AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----', 'Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----', 'NNFS4-----A----', 'AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----', 'NNNS4-----A----', 'NNIS4-----A----', 'CrIS4----------', 'ClIS4----------', 'PDIS4----------', 'AANP1----1A----', 'CrNP1----------', 'AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------', 'AAIS4----1A----'], loss: -0.017212197393538534
['AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------', 'PDFS4----------', 'AAFS4----2A----', 'ClFS4----------', 'P8FS4----------', 'AAFS4----1N----', 'PZFS4----------', 'PWFS4----------', 'PSFS4-P1-------', 'P8FS4---------1', 'AAFS4----1A----', 'RV--4----------', 'RR--4----------'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----', 'PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----', 'AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----', 'Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----', 'NNFS4-----A----', 'AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----', 'NNNS4-----A----', 'NNIS4-----A----', 'CrIS4----------', 'ClIS4----------', 'PDIS4----------', 'AANP1----1A----', 'CrNP1----------', 'AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------', 'AAIS4----1A----', 'P7-X3----------', 'P7-X4----------', 'J^-------------', 'VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---', 'VpYS---XR-AA---', 'VpYS---XR-AA--1', 'VpQW---XR-AA---', 'VpNS---XR-AA---', 'VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---', 'J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---', 'VB-S---3P-AA---', 'PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------', 'PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------', 'AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------', 'Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----', 'VB-S---1P-AA---', 'Vc-S---1-------', 'Vc-X---3-------', 'VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---', 'VB-P---1P-AA---'], loss: 0.014063921010342928
['AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----', 'AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----', 'AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------', 'AAFP1----1A----', 'VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----', 'ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----', 'NNNS2-----A----', 'NNIS2-----A----', 'P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----'] into ['AAFS2----1A----'], loss: 0.021425650217452838
['AAFS2----1A----', 'AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----', 'AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----', 'AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------', 'AAFP1----1A----', 'VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----', 'ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----', 'NNNS2-----A----', 'NNIS2-----A----', 'P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----'] into ['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------', 'PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------', 'AANS1----2A----', 'AANS1----1N----', 'Vf--------N----', 'AAFS1----1N----', 'AAFS1----2A----', 'AAFS1----1A----', 'PDYS1----------', 'PLYS1----------', 'PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------', 'CrIS1----------', 'AAIS1----3A----', 'AAIS1----1A----'], loss: 0.004693733206278717
['AUIS2M---------', 'AAIS2----1N----', 'AAIS2----1A----', 'NNFS2-----A----'] into ['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------', 'PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------', 'AANS1----2A----', 'AANS1----1N----', 'Vf--------N----', 'AAFS1----1N----', 'AAFS1----2A----', 'AAFS1----1A----', 'PDYS1----------', 'PLYS1----------', 'PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------', 'CrIS1----------', 'AAIS1----3A----', 'AAIS1----1A----', 'AAFS2----1A----', 'AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----', 'AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----', 'AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------', 'AAFP1----1A----', 'VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----', 'ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----', 'NNNS2-----A----', 'NNIS2-----A----', 'P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----'], loss: 0.015971563910243278
['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------', 'PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------', 'AANS1----2A----', 'AANS1----1N----', 'Vf--------N----', 'AAFS1----1N----', 'AAFS1----2A----', 'AAFS1----1A----', 'PDYS1----------', 'PLYS1----------', 'PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------', 'CrIS1----------', 'AAIS1----3A----', 'AAIS1----1A----', 'AAFS2----1A----', 'AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----', 'AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----', 'AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------', 'AAFP1----1A----', 'VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----', 'ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----', 'NNNS2-----A----', 'NNIS2-----A----', 'P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----', 'AUIS2M---------', 'AAIS2----1N----', 'AAIS2----1A----', 'NNFS2-----A----'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----', 'PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----', 'AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----', 'Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----', 'NNFS4-----A----', 'AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----', 'NNNS4-----A----', 'NNIS4-----A----', 'CrIS4----------', 'ClIS4----------', 'PDIS4----------', 'AANP1----1A----', 'CrNP1----------', 'AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------', 'AAIS4----1A----', 'P7-X3----------', 'P7-X4----------', 'J^-------------', 'VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---', 'VpYS---XR-AA---', 'VpYS---XR-AA--1', 'VpQW---XR-AA---', 'VpNS---XR-AA---', 'VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---', 'J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---', 'VB-S---3P-AA---', 'PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------', 'PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------', 'AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------', 'Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----', 'VB-S---1P-AA---', 'Vc-S---1-------', 'Vc-X---3-------', 'VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---', 'VB-P---1P-AA---', 'AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------', 'PDFS4----------', 'AAFS4----2A----', 'ClFS4----------', 'P8FS4----------', 'AAFS4----1N----', 'PZFS4----------', 'PWFS4----------', 'PSFS4-P1-------', 'P8FS4---------1', 'AAFS4----1A----', 'RV--4----------', 'RR--4----------'], loss: -0.0014891800435382317
['AAFS2----2A----', 'CrFS2----------', 'AUFS2M---------', 'RV--2----------', 'RR--2----------'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----', 'PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----', 'AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----', 'Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----', 'NNFS4-----A----', 'AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----', 'NNNS4-----A----', 'NNIS4-----A----', 'CrIS4----------', 'ClIS4----------', 'PDIS4----------', 'AANP1----1A----', 'CrNP1----------', 'AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------', 'AAIS4----1A----', 'P7-X3----------', 'P7-X4----------', 'J^-------------', 'VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---', 'VpYS---XR-AA---', 'VpYS---XR-AA--1', 'VpQW---XR-AA---', 'VpNS---XR-AA---', 'VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---', 'J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---', 'VB-S---3P-AA---', 'PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------', 'PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------', 'AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------', 'Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----', 'VB-S---1P-AA---', 'Vc-S---1-------', 'Vc-X---3-------', 'VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---', 'VB-P---1P-AA---', 'AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------', 'PDFS4----------', 'AAFS4----2A----', 'ClFS4----------', 'P8FS4----------', 'AAFS4----1N----', 'PZFS4----------', 'PWFS4----------', 'PSFS4-P1-------', 'P8FS4---------1', 'AAFS4----1A----', 'RV--4----------', 'RR--4----------', 'AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------', 'PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------', 'AANS1----2A----', 'AANS1----1N----', 'Vf--------N----', 'AAFS1----1N----', 'AAFS1----2A----', 'AAFS1----1A----', 'PDYS1----------', 'PLYS1----------', 'PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------', 'CrIS1----------', 'AAIS1----3A----', 'AAIS1----1A----', 'AAFS2----1A----', 'AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----', 'AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----', 'AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------', 'AAFP1----1A----', 'VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----', 'ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----', 'NNNS2-----A----', 'NNIS2-----A----', 'P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----', 'AUIS2M---------', 'AAIS2----1N----', 'AAIS2----1A----', 'NNFS2-----A----'], loss: 0.01619700027927127
['NNMP2-----A----', 'NNNP2-----A----', 'NNIP2-----A----', 'NNFP2-----A----'] into ['AGMP2-----A----', 'AAMP2----1N----', 'AAMP2----1A----', 'AAIP2----1N----', 'AAIP2----2A----', 'AGIP2-----A----', 'AAIP2----1A----', 'AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------', 'CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------', 'AGFP2-----A----', 'AAFP2----1N----', 'PDXP2----------', 'AANP2----1A----', 'PZXP2----------', 'P8XP2----------', 'Ca--2----------', 'AAMP2----3A----', 'PLXP2----------', 'ClXS2----------', 'Ca--4----------', 'Ca--1----------', 'Cn-S1----------', 'Cn-S4----------', 'AAFP2----1A----'], loss: 0.0228845129938635
1: ['AAFS6----3A----'] into ['PDFS6----------']
2: ['AAFS6----1N----'] into ['PSFS6-P1-------']
3: ['PSZS6-P1-------'] into ['CrIS6----------']
4: ['PLXP6----------'] into ['AGFP6-----A----']
5: ['AAIP6----3A----'] into ['AAIP6----1A----']
6: ['CrFS6----------'] into ['PSFS6-P1-------', 'AAFS6----1N----']
7: ['AAIS6----1N----'] into ['PSZS6FS3-------']
8: ['PZXP6----------'] into ['Ca--6----------']
9: ['CrNS6----------'] into ['AUNS6M---------']
10: ['AUFS7M---------'] into ['AAFS7----1N----']
11: ['PJZS3----------'] into ['P4MS4----------']
12: ['J^------------8'] into ['VB-S---2P-NA---']
13: ['AGIS7-----A----'] into ['AAIS7----3A----']
14: ['ClIS4----------'] into ['CrIS4----------']
15: ['ClZS6----------'] into ['PSZS6FS3-------', 'AAIS6----1N----']
16: ['AAFP7----1A----'] into ['AAFP7----2A----']
17: ['AGFS6-----A----'] into ['P4FS6----------']
18: ['PSHS1-S1-------'] into ['PSHS1-P1-------']
19: ['AANS7----1A----'] into ['AANS7----1N----']
20: ['CrFS2----------'] into ['AAFS2----2A----']
21: ['AGFS7-----A----'] into ['AAFS7----3A----']
22: ['PZFS6----------'] into ['ClFS6----------']
23: ['CrNS2----------'] into ['AUNS2M---------']
24: ['Cn-P6----------'] into ['PSXP6-P1-------']
25: ['PZFS1----------'] into ['PWFS1----------']
26: ['AUFS6M---------'] into ['PSFS6-P1-------', 'AAFS6----1N----', 'CrFS6----------']
27: ['AAIP2----2A----'] into ['AAIP2----1N----']
28: ['P5FS2--3-------'] into ['P5ZS2--3------1']
29: ['P4FS7----------'] into ['P9ZS7----------']
30: ['P9XP6----------'] into ['PQ--6----------']
31: ['PSHP1-P2-------'] into ['CrIP1----------']
32: ['NNNSX-----A----'] into ['P6-X6----------']
33: ['AAIP4----1N----'] into ['PZYP4----------']
34: ['AAIS2----2A----'] into ['AAIS2----3A----']
35: ['NNISX-----A----'] into ['AAIS1----1A---6']
36: ['PKM-4----------'] into ['ClHP4----------']
37: ['AUMS1M---------'] into ['AAMS1----1N----']
38: ['PQ--2----------'] into ['PKM-3----------']
39: ['PLIP1----------'] into ['PDIP1----------']
40: ['AGIS4-----A----'] into ['AAIS4----3A----']
41: ['PWFS4----------'] into ['PZFS4----------']
42: ['ACQW------A----'] into ['VB-S---1F-AA---']
43: ['AGNP2-----A----'] into ['CrNP2----------']
44: ['AAMS7----3A----'] into ['AAMS7----1A----']
45: ['CrIS6----------', 'PSZS6-P1-------'] into ['PSZS6FS3-------', 'AAIS6----1N----', 'ClZS6----------']
46: ['AAMP1----2A----'] into ['PSMP1-P1-------']
47: ['PJNS1----------'] into ['PJYS1----------']
48: ['AAIP1----3A----'] into ['AAIP1----1N----']
49: ['CrIS7----------'] into ['VeYS------A----']
50: ['PDFS7----------'] into ['AAFS7----1N----', 'AUFS7M---------']
51: ['P8XP6----------'] into ['AGFP6-----A----', 'PLXP6----------']
52: ['AUIS1M---------'] into ['PWYS1----------']
53: ['P5FS3--3-------'] into ['P5ZS3--3-------']
54: ['AAIP7----3A----'] into ['NNNP1-----A---2']
55: ['ClMS4----------'] into ['P8MS4----------']
56: ['ACTP------A----'] into ['P6-X3----------']
57: ['NNFS2-----N----'] into ['AGNP6-----A----']
58: ['AGFP1-----A----'] into ['CrFP1----------']
59: ['PP-P7--1-------'] into ['P6-X7----------']
60: ['AAFP2----3A----'] into ['AAFP2----2A----']
61: ['PSYS1-S1-------'] into ['PSYS1-P1-------']
62: ['P4NS4----------'] into ['J,-S---1-------']
63: ['Dg-------3A---1'] into ['Dg-------1A---1']
64: ['AGFP7-----A----'] into ['P9ZS7----------', 'P4FS7----------']
65: ['P5XP6--3-------'] into ['P5FS6--3-------']
66: ['PJXP1----------'] into ['P4XP3----------']
67: ['AAFS3----1A----'] into ['PDFS3----------']
68: ['AGNS2-----A----'] into ['NNNS1-----N----']
69: ['PLZS6----------'] into ['AUIS6M---------']
70: ['CrNS4----------'] into ['AANS4----1N----']
71: ['AUFS4M---------'] into ['AAFS4----3A----']
72: ['AAFS4----1N----'] into ['P8FS4----------']
73: ['P4YP4----------'] into ['P4MS4----------', 'PJZS3----------']
74: ['PZIS4----------'] into ['AAIS4----1N----']
75: ['P8XP7----------'] into ['Cn-P7----------']
76: ['ClFS2----------'] into ['AAFS2----1N----']
77: ['PDXP6----------'] into ['Ca--6----------', 'PZXP6----------']
78: ['AAIS2----1N----'] into ['AUIS2M---------']
79: ['VB-S---1P-NA--1'] into ['VB-P---1F-AA---']
80: ['PP-P1--2-------'] into ['AAIP1----2A----']
81: ['PSZS2-S1-------'] into ['PSZS2FS3-------']
82: ['AAFP4----1N----'] into ['PSXP4-P1-------']
83: ['AAMP1----3A----'] into ['AGMP1-----A----']
84: ['PZZS7----------'] into ['PLNS4---------1']
85: ['PZM-4----------'] into ['VB-S---1F-AA---', 'ACQW------A----']
86: ['AAMS6----1A----'] into ['NNMS6-----A---1']
87: ['Co-------------'] into ['Dg-------1A---3']
88: ['AAFP1----3A----'] into ['PLFP1----------']
89: ['P4XP2----------'] into ['P6-X2----------']
90: ['AANS1----1N----'] into ['AANS1----2A----']
91: ['PQ--7----------'] into ['J^------------1']
92: ['PLXP3----------'] into ['ClXP3----------']
93: ['NNFP1-----N----'] into ['NNMP6-----A----']
94: ['J,-P---1-------'] into ['Vi-P---2--N----']
95: ['AGIS2-----A----'] into ['ClZS3----------']
96: ['Vi-S---2--A----'] into ['VB-S---2P-NA---', 'J^------------8']
97: ['AAFP1----1N----'] into ['AAFP1----2A----']
98: ['NNIP2-----A---1'] into ['NNFS1-----N----']
99: ['P9XP2----------'] into ['P5ZS2--3------1', 'P5FS2--3-------']
100: ['AAXXX----1A---8'] into ['A2--------A----']
101: ['NNFP2-----A---1'] into ['NNIS5-----A----']
102: ['CyFP1----------'] into ['AAIS1----1A---6', 'NNISX-----A----']
103: ['AAMS1----2A----'] into ['CrMS1----------']
104: ['P9ZS6----------'] into ['PQ--6----------', 'P9XP6----------']
105: ['PDXP7----------'] into ['AAIS7----1N----']
106: ['II-------------'] into ['NNFS5-----A----']
107: ['C?--4----------'] into ['NNNS1-----N----', 'AGNS2-----A----']
108: ['CrNP1----------'] into ['AANP1----1A----']
109: ['CrIP1----------', 'PSHP1-P2-------'] into ['AAIP1----1N----', 'AAIP1----3A----']
110: ['P8FP4---------1'] into ['PLFP4----------']
111: ['VB-P---3F-NA---'] into ['VB-P---3P-NA--1']
112: ['PLNS1---------1'] into ['ClNS1----------']
113: ['PSZS1FS3-------'] into ['PZYS1----------']
114: ['P6-X4----------'] into ['P5ZS4--3-------']
115: ['NNFS2-----A---1'] into ['AGNP6-----A----', 'NNFS2-----N----']
116: ['AAFS7----3A----', 'AGFS7-----A----'] into ['AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------']
117: ['PDMS4----------'] into ['VB-P---2P-NA---']
118: ['PSFS4-P1-------'] into ['PZFS4----------', 'PWFS4----------']
119: ['CrFS4----------'] into ['AAFS4----3A----', 'AUFS4M---------']
120: ['AAIS1----1N----'] into ['AAIS1----2A----']
121: ['PSXP2-P1-------'] into ['AAFP2----2A----', 'AAFP2----3A----']
122: ['P5XP7--3-------'] into ['P5ZS7--3-------']
123: ['PDIP4----------'] into ['PZYP4----------', 'AAIP4----1N----']
124: ['P5ZS6--3-------'] into ['P9FS6----------']
125: ['PPXP2--3-------'] into ['PPZS7--3-------']
126: ['ClXP6----------'] into ['PSXP6-P1-------', 'Cn-P6----------']
127: ['PP-P2--1-------'] into ['P5XP2--3-------']
128: ['AGIP2-----A----'] into ['AAIP2----1N----', 'AAIP2----2A----']
129: ['P4ZS6----------'] into ['P8ZS6----------']
130: ['PZFP1----------'] into ['PSHP1-P1-------']
131: ['PZMP1----------'] into ['PPMP1--3-------']
132: ['AAMP2----1N----'] into ['AGMP2-----A----']
133: ['PZFP4----------'] into ['PWFP4----------']
134: ['AANS1----3A----'] into ['PLNS4----------']
135: ['P5FS4--3-------'] into ['P5XP4--3-------']
136: ['RR--X----------'] into ['RR--1----------']
137: ['P4ZS7----------'] into ['PKM-3----------', 'PQ--2----------']
138: ['ClFS4----------'] into ['AAFS4----2A----']
139: ['P5ZS2--3-------'] into ['P6-X2----------', 'P4XP2----------']
140: ['AANS2----1N----'] into ['AUNS2M---------', 'CrNS2----------']
141: ['AUFS2M---------'] into ['AAFS2----2A----', 'CrFS2----------']
142: ['Vf--------A---1'] into ['NNNP1-----A---2', 'AAIP7----3A----']
143: ['PJYS1----------', 'PJNS1----------'] into ['J,-S---1-------', 'P4NS4----------']
144: ['AAIS1----3A----'] into ['CrIS1----------']
145: ['AGIS6-----A----'] into ['ClFS6----------', 'PZFS6----------']
146: ['PSHS1-P1-------', 'PSHS1-S1-------'] into ['PWFS1----------', 'PZFS1----------']
147: ['Vi-P---1--A----'] into ['Vi-P---2--A----']
148: ['ClZS2----------'] into ['CrIS2----------']
149: ['P6-X6----------', 'NNNSX-----A----'] into ['P5FS6--3-------', 'P5XP6--3-------']
150: ['ClXP7----------'] into ['Cn-P7----------', 'P8XP7----------']
151: ['AGMS1-----A----'] into ['AAMS1----1N----', 'AUMS1M---------']
152: ['PLFS1----------'] into ['P6-X3----------', 'ACTP------A----']
153: ['PDMP4----------'] into ['PLNS4---------1', 'PZZS7----------']
154: ['PLXP7----------'] into ['NNMP2-----A---1']
155: ['PWYS1----------', 'AUIS1M---------'] into ['PZYS1----------', 'PSZS1FS3-------']
156: ['NNIXX-----A---8'] into ['NNMXX-----A---8']
157: ['P4FP1----------'] into ['P4IP1----------']
158: ['P8FS6---------1'] into ['P4FS6----------', 'AGFS6-----A----']
159: ['AUIS6M---------', 'PLZS6----------'] into ['P8ZS6----------', 'P4ZS6----------']
160: ['AAIS7----3A----', 'AGIS7-----A----'] into ['AAIS7----1N----', 'PDXP7----------']
161: ['PPNS4--3-------'] into ['P8ZS3----------']
162: ['AGFS2-----A----'] into ['PSFS2-P1-------']
163: ['AAFS6----1A----'] into ['PSFS6-P1-------', 'AAFS6----1N----', 'CrFS6----------', 'AUFS6M---------']
164: ['P4FS4----------'] into ['P4MS4----------', 'PJZS3----------', 'P4YP4----------']
165: ['AAMP3----1A----'] into ['P8XP3----------']
166: ['ClFS1----------'] into ['AAFS1----3A----']
167: ['P8YP4---------1'] into ['PLYP4----------']
168: ['AAIS2----3A----', 'AAIS2----2A----'] into ['CrIS2----------', 'ClZS2----------']
169: ['AUFS1M---------'] into ['NNFS1-----A---1']
170: ['PSMP1-P1-------', 'AAMP1----2A----'] into ['AGMP1-----A----', 'AAMP1----3A----']
171: ['AAFP2----1N----'] into ['AGFP2-----A----']
172: ['AAMP1----1N----'] into ['CrFP1----------', 'AGFP1-----A----']
173: ['AAMP2----3A----'] into ['Ca--2----------']
174: ['PDFP4----------'] into ['ClHP4----------', 'PKM-4----------']
175: ['P1XXXXP3-------'] into ['P1XXXZS3-------']
176: ['Dg-------1A---1', 'Dg-------3A---1'] into ['Dg-------1A---3', 'Co-------------']
177: ['ClZS7----------'] into ['ClFS7----------']
178: ['Cn-P2----------'] into ['CrNP2----------', 'AGNP2-----A----']
179: ['PP-P4--2-------'] into ['PP-S4--1-------']
180: ['P4IS4----------'] into ['P4FP4----------']
181: ['PP-P1--1-------'] into ['AAIP1----2A----', 'PP-P1--2-------']
182: ['AAMS4----1A----'] into ['P8MS4----------', 'ClMS4----------']
183: ['P8FS4----------', 'AAFS4----1N----'] into ['AAFS4----2A----', 'ClFS4----------']
184: ['AUNS6M---------', 'CrNS6----------'] into ['PSZS6FS3-------', 'AAIS6----1N----', 'ClZS6----------', 'CrIS6----------', 'PSZS6-P1-------']
185: ['AAFS1----2A----'] into ['AAFS1----1N----']
186: ['CrNS1----------'] into ['AUNS1M---------']
187: ['P8NS4---------1'] into ['AANS4----1N----', 'CrNS4----------']
188: ['NNFS4-----N----'] into ['AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1']
189: ['NNIS2-----A---1'] into ['ClNS1----------', 'PLNS1---------1']
190: ['P8IS4----------'] into ['AAIS4----3A----', 'AGIS4-----A----']
191: ['P8XP2----------'] into ['PZXP2----------']
192: ['PJYS2----------'] into ['Vi-P---2--N----', 'J,-P---1-------']
193: ['VpTP---XR-AA--1'] into ['VpNS---XR-AA--1']
194: ['PDXP3----------'] into ['ClXP3----------', 'PLXP3----------']
195: ['P4FS1----------'] into ['PE--1----------']
196: ['PDIP1----------', 'PLIP1----------'] into ['AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------']
197: ['VB-P---1F-AA---', 'VB-S---1P-NA--1'] into ['VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------']
198: ['PPYS1--3-------'] into ['AGIP1-----A----']
199: ['AANP7----1A----'] into ['NNIS5-----A----', 'NNFP2-----A---1']
200: ['P8FS2---------1'] into ['AAFS2----1N----', 'ClFS2----------']
201: ['VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----'] into ['NNFS5-----A----', 'II-------------']
202: ['ClZS3----------', 'AGIS2-----A----'] into ['VB-P---2P-NA---', 'PDMS4----------']
203: ['P4NS1----------'] into ['P4XP3----------', 'PJXP1----------']
204: ['PPFS3--3-------'] into ['PZ--1----------']
205: ['CrMS1----------', 'AAMS1----2A----'] into ['AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----']
206: ['CrFS1----------'] into ['PSFSXFS3-------']
207: ['P9FS6----------', 'P5ZS6--3-------'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----']
208: ['PSXP4-P1-------', 'AAFP4----1N----'] into ['PLNS4---------1', 'PZZS7----------', 'PDMP4----------']
209: ['PDIS4----------'] into ['CrIS4----------', 'ClIS4----------']
210: ['P8FS7---------1'] into ['AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----']
211: ['NNNP3-----A----'] into ['NNFS1-----N----', 'NNIP2-----A---1']
212: ['PWFP4----------', 'PZFP4----------'] into ['ClHP4----------', 'PKM-4----------', 'PDFP4----------']
213: ['ClHP1----------'] into ['PLFP1----------', 'AAFP1----3A----']
214: ['NNMS6-----A---1', 'AAMS6----1A----'] into ['NNMP6-----A----', 'NNFP1-----N----']
215: ['VB-S---2P-AA---'] into ['AGNS1-----A----']
216: ['PSZS2-P1-------'] into ['P8ZS2----------']
217: ['PSZS2FS3-------', 'PSZS2-S1-------'] into ['P8ZS2----------', 'PSZS2-P1-------']
218: ['VeYS------A----', 'CrIS7----------'] into ['AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----']
219: ['AAIS2----1A----'] into ['AUIS2M---------', 'AAIS2----1N----']
220: ['P8FS4---------1'] into ['PZFS4----------', 'PWFS4----------', 'PSFS4-P1-------']
221: ['AAIS4----1N----', 'PZIS4----------'] into ['AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------']
222: ['ClYP4----------'] into ['ClXP4----------']
223: ['PLNS4----------', 'AANS1----3A----'] into ['AGNS1-----A----', 'VB-S---2P-AA---']
224: ['ClYS1----------'] into ['AGIS1-----A----']
225: ['PW--1----------'] into ['PW--4----------']
226: ['NNMS3-----A---1'] into ['AAMS3----1A----']
227: ['VB-S---3F-NA---'] into ['J^------------1', 'PQ--7----------']
228: ['P5ZS3--3-------', 'P5FS3--3-------'] into ['P6-X3----------', 'ACTP------A----', 'PLFS1----------']
229: ['Vc-P---1-------'] into ['ACMP------A----']
230: ['PZM-1----------'] into ['PP-P3--2-------']
231: ['P4YS1----------'] into ['J,-S---1-------', 'P4NS4----------', 'PJYS1----------', 'PJNS1----------']
232: ['VB-P---1P-NA---'] into ['VB-S---1P-AA--1']
233: ['PPMP1--3-------', 'PZMP1----------'] into ['AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----']
234: ['RF-------------'] into ['NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------']
235: ['P5ZS4--3-------', 'P6-X4----------'] into ['P5XP4--3-------', 'P5FS4--3-------']
236: ['VpTP---XR-NA---'] into ['VB-P---3P-NA--1', 'VB-P---3F-NA---']
237: ['PPZS7--3-------', 'PPXP2--3-------'] into ['P8ZS3----------', 'PPNS4--3-------']
238: ['CrIS1----------', 'AAIS1----3A----'] into ['PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------']
239: ['PQ--1----------'] into ['P4XP3----------', 'PJXP1----------', 'P4NS1----------']
240: ['A2--------A----', 'AAXXX----1A---8'] into ['NNMXX-----A---8', 'NNIXX-----A---8']
241: ['AANS2----1A----'] into ['AUNS2M---------', 'CrNS2----------', 'AANS2----1N----']
242: ['NNMS5-----A----'] into ['AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------']
243: ['NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1'] into ['CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----']
244: ['PDFS1----------'] into ['PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------']
245: ['PSYS1-P1-------', 'PSYS1-S1-------'] into ['AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----']
246: ['ClXP2----------'] into ['AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------']
247: ['PDFS2----------'] into ['PSFS2-P1-------', 'AGFS2-----A----']
248: ['PKM-1----------'] into ['PKM-3----------', 'PQ--2----------', 'P4ZS7----------']
249: ['Ca--2----------', 'AAMP2----3A----'] into ['PZXP2----------', 'P8XP2----------']
250: ['PH-S4--1-------'] into ['Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1']
251: ['AAIP2----1A----'] into ['AAIP2----1N----', 'AAIP2----2A----', 'AGIP2-----A----']
252: ['AANP6----1A----'] into ['PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------']
253: ['Ca--1----------'] into ['Ca--4----------']
254: ['PDFS4----------'] into ['AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------']
255: ['AAIP4----1A----'] into ['PZYP4----------', 'AAIP4----1N----', 'PDIP4----------']
256: ['NNMS6-----A----'] into ['NNMP6-----A----', 'NNFP1-----N----', 'NNMS6-----A---1', 'AAMS6----1A----']
257: ['PZ--4----------'] into ['ACYS------A----']
258: ['Vf--------N----'] into ['AANS1----2A----', 'AANS1----1N----']
259: ['P8ZS7----------'] into ['AANP3----1A----']
260: ['AAMP7----1A----'] into ['P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----']
261: ['PP-P4--1-------'] into ['PP-S4--1-------', 'PP-P4--2-------']
262: ['RR--1----------', 'RR--X----------'] into ['NNFS1-----A---1', 'AUFS1M---------']
263: ['AGFS1-----A----'] into ['AAFS1----3A----', 'ClFS1----------']
264: ['PSHP1-P1-------', 'PZFP1----------'] into ['PLFP1----------', 'AAFP1----3A----', 'ClHP1----------']
265: ['VpQW---XR-NA---'] into ['VpQW---XR-AA--1']
266: ['PQ--6----------', 'P9XP6----------', 'P9ZS6----------'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----', 'P9FS6----------', 'P5ZS6--3-------']
267: ['AAFS4----2A----', 'ClFS4----------', 'P8FS4----------', 'AAFS4----1N----'] into ['AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------', 'PDFS4----------']
268: ['PPFS4--3-------'] into ['PHZS4--3-------']
269: ['ClXP1----------'] into ['AAFP1----2A----', 'AAFP1----1N----']
270: ['NNFSX-----A----'] into ['NNIPX-----A----']
271: ['NNMP2-----A---1', 'PLXP7----------'] into ['AGIP1-----A----', 'PPYS1--3-------']
272: ['PDMP1----------'] into ['PLMP1----------']
273: ['PDXP2----------'] into ['AGFP2-----A----', 'AAFP2----1N----']
274: ['PWM-1----------'] into ['PLNS1----------']
275: ['AAFS7----1A----'] into ['AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1']
276: ['PLFP4----------', 'P8FP4---------1'] into ['ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------']
277: ['PP-S1--1-------'] into ['AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------']
278: ['PPXP4--3-------'] into ['PPZS4--3------2']
279: ['VpMP---XR-AA--1'] into ['VpMP---XR-NA---']
280: ['PLXP2----------'] into ['PZXP2----------', 'P8XP2----------', 'Ca--2----------', 'AAMP2----3A----']
281: ['VB-P---3P-AA--1'] into ['VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1']
282: ['AANP4----1A----'] into ['PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----']
283: ['AANS4----1A----'] into ['AANS4----1N----', 'CrNS4----------', 'P8NS4---------1']
284: ['Vi-P---2--A----', 'Vi-P---1--A----'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----']
285: ['PSXXXXP3-------'] into ['P1XXXZS3-------', 'P1XXXXP3-------']
286: ['AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1'] into ['PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------']
287: ['AAMP2----1A----'] into ['AGMP2-----A----', 'AAMP2----1N----']
288: ['NNNP7-----A----'] into ['P6-X7----------', 'PP-P7--1-------']
289: ['P5ZS7--3-------', 'P5XP7--3-------'] into ['P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----']
290: ['Db------------8'] into ['NNFXX-----A---8']
291: ['Dg-------3A----'] into ['ClFS7----------', 'ClZS7----------']
292: ['P4MS4----------', 'PJZS3----------', 'P4YP4----------', 'P4FS4----------'] into ['Vi-P---2--N----', 'J,-P---1-------', 'PJYS2----------']
293: ['P4FP4----------', 'P4IS4----------'] into ['Vi-P---2--N----', 'J,-P---1-------', 'PJYS2----------', 'P4MS4----------', 'PJZS3----------', 'P4YP4----------', 'P4FS4----------']
294: ['AGFS4-----A----'] into ['VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----']
295: ['PHZS3--3-------'] into ['PZ--1----------', 'PPFS3--3-------']
296: ['Cn-P7----------', 'P8XP7----------', 'ClXP7----------'] into ['P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----']
297: ['PPXP3--3-------'] into ['PP-P3--2-------', 'PZM-1----------']
298: ['P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1'] into ['P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------']
299: ['ClFS6----------', 'PZFS6----------', 'AGIS6-----A----'] into ['P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1']
300: ['AAIP6----1A----', 'AAIP6----3A----'] into ['Ca--6----------', 'PZXP6----------', 'PDXP6----------']
301: ['P4MP1----------'] into ['P4IP1----------', 'P4FP1----------']
302: ['VsMP---XX-AP---'] into ['ACMP------A----', 'Vc-P---1-------']
303: ['AAFP6----1A----'] into ['AGFP6-----A----', 'PLXP6----------', 'P8XP6----------']
304: ['P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------'] into ['ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1']
305: ['ClXP3----------', 'PLXP3----------', 'PDXP3----------'] into ['P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------']
306: ['AAIS4----1A----'] into ['AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------']
307: ['C}-------------'] into ['NNMXX-----A---8', 'NNIXX-----A---8', 'A2--------A----', 'AAXXX----1A---8']
308: ['AAIS1----2A----', 'AAIS1----1N----'] into ['AGIS1-----A----', 'ClYS1----------']
309: ['P5XP2--3-------', 'PP-P2--1-------'] into ['ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------']
310: ['ClYP1----------'] into ['AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------']
311: ['AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------'] into ['PLMP1----------', 'PDMP1----------']
312: ['PLYS1----------'] into ['PDYS1----------']
313: ['PDFP1----------'] into ['PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------']
314: ['VB-P---2P-AA---'] into ['VB-S---1P-NA---']
315: ['PSFSXFS3-------', 'CrFS1----------'] into ['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----']
316: ['Vc-S---1-------'] into ['VB-S---1P-AA---']
317: ['VpNS---XR-AA--1', 'VpTP---XR-AA--1'] into ['VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---']
318: ['PP-P3--1-------'] into ['PH-S3--1-------']
319: ['NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----'] into ['J^------------1', 'PQ--7----------', 'VB-S---3F-NA---']
320: ['AANS6----1A----'] into ['PSZS6FS3-------', 'AAIS6----1N----', 'ClZS6----------', 'CrIS6----------', 'PSZS6-P1-------', 'AUNS6M---------', 'CrNS6----------']
321: ['PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------', 'CrIS1----------', 'AAIS1----3A----'] into ['PDYS1----------', 'PLYS1----------']
322: ['PDZS7----------'] into ['AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----', 'VeYS------A----', 'CrIS7----------']
323: ['AANS1----1A----'] into ['AUNS1M---------', 'CrNS1----------']
324: ['AAMP4----1A----'] into ['PLYP4----------', 'P8YP4---------1']
325: ['CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----'] into ['AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----']
326: ['Dg-------2A---1'] into ['Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------']
327: ['Db------------1'] into ['Cv-------------']
328: ['P8XP3----------', 'AAMP3----1A----'] into ['P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------']
329: ['NNFPX-----A----'] into ['AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------', 'NNMS5-----A----']
330: ['ACNS------A----'] into ['AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----']
331: ['VB-S---1P-AA--1', 'VB-P---1P-NA---'] into ['VB-S---1P-NA---', 'VB-P---2P-AA---']
332: ['CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------'] into ['AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------']
333: ['AAFS1----1N----', 'AAFS1----2A----'] into ['AANS1----2A----', 'AANS1----1N----', 'Vf--------N----']
334: ['Cn-S1----------'] into ['Ca--4----------', 'Ca--1----------']
335: ['CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1'] into ['AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------']
336: ['PDZS2----------'] into ['P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------']
337: ['PE--1----------', 'P4FS1----------'] into ['J,-S---1-------', 'P4NS4----------', 'PJYS1----------', 'PJNS1----------', 'P4YS1----------']
338: ['P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------'] into ['PPZS4--3------2', 'PPXP4--3-------']
339: ['AAIP3----1A----'] into ['AANP3----1A----', 'P8ZS7----------']
340: ['PW--4----------', 'PW--1----------'] into ['PLNS1----------', 'PWM-1----------']
341: ['NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------'] into ['AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------']
342: ['Dg-------1N----'] into ['ACYS------A----', 'PZ--4----------']
343: ['NNMS3-----A----'] into ['AAMS3----1A----', 'NNMS3-----A---1']
344: ['AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----'] into ['VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----']
345: ['VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----', 'AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----'] into ['AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----', 'CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----']
346: ['AAMS1----1A----'] into ['AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----', 'PSYS1-P1-------', 'PSYS1-S1-------']
347: ['PZFS4----------', 'PWFS4----------', 'PSFS4-P1-------', 'P8FS4---------1'] into ['AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------', 'PDFS4----------', 'AAFS4----2A----', 'ClFS4----------', 'P8FS4----------', 'AAFS4----1N----']
348: ['NNIS6-----A----'] into ['NNIS6-----A---1']
349: ['NNMP1-----A----'] into ['NNMP1-----A---1']
350: ['AAMP1----1A----'] into ['PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------']
351: ['PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----'] into ['AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----']
352: ['NNFP6-----A----'] into ['NNNP6-----A----']
353: ['ClXP4----------', 'ClYP4----------'] into ['PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----']
354: ['PQ--4----------'] into ['PKM-3----------', 'PQ--2----------', 'P4ZS7----------', 'PKM-1----------']
355: ['RV--3----------'] into ['RR--3----------']
356: ['VpYS---XR-AA--1'] into ['VpYS---XR-AA---']
357: ['AAFP3----1A----'] into ['NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----']
358: ['AGFP2-----A----', 'AAFP2----1N----', 'PDXP2----------'] into ['AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------', 'CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------']
359: ['VB-P---3P-NA---'] into ['VpMP---XR-NA---', 'VpMP---XR-AA--1']
360: ['VpTP---XR-AA---'] into ['VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1']
361: ['ClXS2----------'] into ['PZXP2----------', 'P8XP2----------', 'Ca--2----------', 'AAMP2----3A----', 'PLXP2----------']
362: ['Cn-S4----------'] into ['Ca--4----------', 'Ca--1----------', 'Cn-S1----------']
363: ['PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------'] into ['PH-S3--1-------', 'PP-P3--1-------']
364: ['NNIPX-----A----', 'NNFSX-----A----'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----', 'Vi-P---2--A----', 'Vi-P---1--A----']
365: ['PHZS4--3-------', 'PPFS4--3-------'] into ['PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------']
366: ['VB-P---3F-AA---'] into ['VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1']
367: ['NNFS1-----A---1', 'AUFS1M---------', 'RR--1----------', 'RR--X----------'] into ['AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------', 'NNMS5-----A----', 'NNFPX-----A----']
368: ['Vi-P---2--N----', 'J,-P---1-------', 'PJYS2----------', 'P4MS4----------', 'PJZS3----------', 'P4YP4----------', 'P4FS4----------', 'P4FP4----------', 'P4IS4----------'] into ['PKM-3----------', 'PQ--2----------', 'P4ZS7----------', 'PKM-1----------', 'PQ--4----------']
369: ['RR--4----------'] into ['RV--4----------']
370: ['AAFP4----1A----'] into ['ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1']
371: ['PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------'] into ['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------']
372: ['VpNS---XR-NA---'] into ['J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----']
373: ['NNFXX-----A---8', 'Db------------8'] into ['NNMXX-----A---8', 'NNIXX-----A---8', 'A2--------A----', 'AAXXX----1A---8', 'C}-------------']
374: ['PDZS6----------'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----', 'P9FS6----------', 'P5ZS6--3-------', 'PQ--6----------', 'P9XP6----------', 'P9ZS6----------']
375: ['NNNP4-----A----'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------']
376: ['NNMP7-----A----'] into ['P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----', 'P5ZS7--3-------', 'P5XP7--3-------']
377: ['PDZS3----------'] into ['AAIS3----1A----']
378: ['NNMS4-----A----'] into ['P8MS4----------', 'ClMS4----------', 'AAMS4----1A----']
379: ['Dg-------2A----'] into ['Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1']
380: ['PDNS4----------'] into ['PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------']
381: ['AAFP7----2A----', 'AAFP7----1A----'] into ['P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----', 'Cn-P7----------', 'P8XP7----------', 'ClXP7----------']
382: ['P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------'] into ['P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------']
383: ['RV--2----------'] into ['AAFS2----2A----', 'CrFS2----------', 'AUFS2M---------']
384: ['J,-X---3-------'] into ['P4XP3----------', 'PJXP1----------', 'P4NS1----------', 'PQ--1----------']
385: ['AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------'] into ['AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------']
386: ['NNFP3-----A----'] into ['NNIP3-----A----']
387: ['NNIP6-----A----'] into ['NNMP6-----A----', 'NNFP1-----N----', 'NNMS6-----A---1', 'AAMS6----1A----', 'NNMS6-----A----']
388: ['ClFS7----------', 'ClZS7----------', 'Dg-------3A----'] into ['AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----']
389: ['NNFP7-----A----'] into ['P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----', 'P5ZS7--3-------', 'P5XP7--3-------', 'NNMP7-----A----']
390: ['AAFS4----1A----'] into ['AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------', 'PDFS4----------', 'AAFS4----2A----', 'ClFS4----------', 'P8FS4----------', 'AAFS4----1N----', 'PZFS4----------', 'PWFS4----------', 'PSFS4-P1-------', 'P8FS4---------1']
391: ['VsYS---XX-AP---'] into ['AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----']
392: ['PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------'] into ['AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1']
393: ['AAIP1----1A----'] into ['AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------']
394: ['RV--7----------'] into ['RR--7----------']
395: ['P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----'] into ['AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----']
396: ['NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----'] into ['AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----']
397: ['NNMP3-----A----'] into ['NNIP3-----A----', 'NNFP3-----A----']
398: ['VB-S---3F-AA---'] into ['VpQW---XR-AA--1', 'VpQW---XR-NA---']
399: ['VB-S---3P-NA---'] into ['VpYS---XR-NA---']
400: ['ACYS------A----', 'PZ--4----------', 'Dg-------1N----'] into ['Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----']
401: ['Db-------------'] into ['Cv-------------', 'Db------------1']
402: ['PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------'] into ['PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------']
403: ['PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------'] into ['PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------']
404: ['VsTP---XX-AP---'] into ['ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---']
405: ['AANS7----1N----', 'AANS7----1A----'] into ['AAMS7----1A----', 'AAMS7----3A----']
406: ['XX-------------'] into ['AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------', 'NNMS5-----A----', 'NNFPX-----A----', 'NNFS1-----A---1', 'AUFS1M---------', 'RR--1----------', 'RR--X----------']
407: ['PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----']
408: ['VB-P---1P-AA---'] into ['VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---']
409: ['VpMP---XR-AA---'] into ['VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---']
410: ['AAIS7----1A----'] into ['AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----', 'VeYS------A----', 'CrIS7----------', 'PDZS7----------']
411: ['VsQW---XX-AP---'] into ['VsNS---XX-AP---']
412: ['PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----'] into ['PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------']
413: ['NNIP4-----A----'] into ['NNMP4-----A----']
414: ['NNNXX-----A----'] into ['NNIXX-----A----']
415: ['PDFS6----------', 'AAFS6----3A----'] into ['Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----']
416: ['NNNS3-----A----'] into ['AANS3----1A----']
417: ['AAIP7----1A----'] into ['P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----', 'Cn-P7----------', 'P8XP7----------', 'ClXP7----------', 'AAFP7----2A----', 'AAFP7----1A----']
418: ['P4IP1----------', 'P4FP1----------', 'P4MP1----------'] into ['P4XP3----------', 'PJXP1----------', 'P4NS1----------', 'PQ--1----------', 'J,-X---3-------']
419: ['AAXXX----1A----'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----', 'Vi-P---2--A----', 'Vi-P---1--A----', 'NNIPX-----A----', 'NNFSX-----A----']
420: ['AANS1----2A----', 'AANS1----1N----', 'Vf--------N----', 'AAFS1----1N----', 'AAFS1----2A----'] into ['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------', 'PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------']
421: ['AANP1----1A----', 'CrNP1----------'] into ['CrIS4----------', 'ClIS4----------', 'PDIS4----------']
422: ['VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---'] into ['VpYS---XR-NA---', 'VB-S---3P-NA---']
423: ['VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---'] into ['J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---']
424: ['NNMS2-----A----'] into ['AAMS2----1A----']
425: ['AAMS2----1A----', 'NNMS2-----A----'] into ['PDFS3----------', 'AAFS3----1A----']
426: ['PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----'] into ['P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------']
427: ['P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----'] into ['P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----']
428: ['P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----'] into ['Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----']
429: ['Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----'] into ['AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----']
430: ['AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----', 'Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----', 'PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----']
431: ['NNNP6-----A----', 'NNFP6-----A----'] into ['NNMP6-----A----', 'NNFP1-----N----', 'NNMS6-----A---1', 'AAMS6----1A----', 'NNMS6-----A----', 'NNIP6-----A----']
432: ['RV--6----------'] into ['RR--6----------']
433: ['RR--2----------'] into ['AAFS2----2A----', 'CrFS2----------', 'AUFS2M---------', 'RV--2----------']
434: ['ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------'] into ['PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------']
435: ['J,-S---1-------', 'P4NS4----------', 'PJYS1----------', 'PJNS1----------', 'P4YS1----------', 'PE--1----------', 'P4FS1----------'] into ['PKM-3----------', 'PQ--2----------', 'P4ZS7----------', 'PKM-1----------', 'PQ--4----------', 'Vi-P---2--N----', 'J,-P---1-------', 'PJYS2----------', 'P4MS4----------', 'PJZS3----------', 'P4YP4----------', 'P4FS4----------', 'P4FP4----------', 'P4IS4----------']
436: ['AAFP1----1A----'] into ['AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------']
437: ['Xx-------------'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----', 'Vi-P---2--A----', 'Vi-P---1--A----', 'NNIPX-----A----', 'NNFSX-----A----', 'AAXXX----1A----']
438: ['AAIS1----1A----'] into ['PDYS1----------', 'PLYS1----------', 'PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------', 'CrIS1----------', 'AAIS1----3A----']
439: ['TT-------------'] into ['AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----']
440: ['AANP2----1A----'] into ['AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------', 'CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------', 'AGFP2-----A----', 'AAFP2----1N----', 'PDXP2----------']
441: ['PZXP2----------', 'P8XP2----------', 'Ca--2----------', 'AAMP2----3A----', 'PLXP2----------', 'ClXS2----------'] into ['AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------', 'CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------', 'AGFP2-----A----', 'AAFP2----1N----', 'PDXP2----------', 'AANP2----1A----']
442: ['X@-------------'] into ['AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------', 'NNMS5-----A----', 'NNFPX-----A----', 'NNFS1-----A---1', 'AUFS1M---------', 'RR--1----------', 'RR--X----------', 'XX-------------']
443: ['VB-P---3P-AA---'] into ['VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---']
444: ['NNFS3-----A----'] into ['NNIP3-----A----', 'NNFP3-----A----', 'NNMP3-----A----']
445: ['Dg-------1A----'] into ['Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----']
446: ['AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------', 'AAIS4----1A----'] into ['CrIS4----------', 'ClIS4----------', 'PDIS4----------', 'AANP1----1A----', 'CrNP1----------']
447: ['AAFS1----1A----'] into ['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------', 'PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------', 'AANS1----2A----', 'AANS1----1N----', 'Vf--------N----', 'AAFS1----1N----', 'AAFS1----2A----']
448: ['P7-X4----------'] into ['P7-X3----------']
449: ['NNNS7-----A----'] into ['NNMS7-----A----']
450: ['NNNP1-----A----'] into ['NNFP1-----A----']
451: ['AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----', 'CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----', 'VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----', 'AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----'] into ['NNIP7-----A----']
452: ['AAIS3----1A----', 'PDZS3----------'] into ['AANS3----1A----', 'NNNS3-----A----']
453: ['P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----'] into ['AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----']
454: ['AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---'] into ['VsNS---XX-AP---', 'VsQW---XX-AP---']
455: ['NNFXX-----A----'] into ['NNIXX-----A----', 'NNNXX-----A----']
456: ['VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---'] into ['J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---']
457: ['PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------'] into ['PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------']
458: ['AAIS6----1A----'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----', 'P9FS6----------', 'P5ZS6--3-------', 'PQ--6----------', 'P9XP6----------', 'P9ZS6----------', 'PDZS6----------']
459: ['NNIS3-----A----'] into ['AANS3----1A----', 'NNNS3-----A----', 'AAIS3----1A----', 'PDZS3----------']
460: ['NNNP2-----A----'] into ['NNMP2-----A----']
461: ['AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------'] into ['PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------']
462: ['Ca--4----------', 'Ca--1----------', 'Cn-S1----------', 'Cn-S4----------'] into ['AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------', 'CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------', 'AGFP2-----A----', 'AAFP2----1N----', 'PDXP2----------', 'AANP2----1A----', 'PZXP2----------', 'P8XP2----------', 'Ca--2----------', 'AAMP2----3A----', 'PLXP2----------', 'ClXS2----------']
463: ['Cv-------------', 'Db------------1', 'Db-------------'] into ['AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------']
464: ['Vf--------A----'] into ['ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---']
465: ['VpNS---XR-AA---'] into ['VpQW---XR-AA---']
466: ['PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1'] into ['P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------']
467: ['P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----', 'P5ZS7--3-------', 'P5XP7--3-------', 'NNMP7-----A----', 'NNFP7-----A----'] into ['NNIP7-----A----', 'AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----', 'CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----', 'VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----', 'AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----']
468: ['Vc-X---3-------'] into ['VB-S---1P-AA---', 'Vc-S---1-------']
469: ['PDNS1----------'] into ['PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------']
470: ['NNFP4-----A----'] into ['NNMP4-----A----', 'NNIP4-----A----']
471: ['AAMS7----1A----', 'AAMS7----3A----', 'AANS7----1N----', 'AANS7----1A----'] into ['P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----', 'Cn-P7----------', 'P8XP7----------', 'ClXP7----------', 'AAFP7----2A----', 'AAFP7----1A----', 'AAIP7----1A----']
472: ['NNMS7-----A----', 'NNNS7-----A----'] into ['NNIP7-----A----', 'AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----', 'CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----', 'VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----', 'AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----', 'P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----', 'P5ZS7--3-------', 'P5XP7--3-------', 'NNMP7-----A----', 'NNFP7-----A----']
473: ['NNMXX-----A---8', 'NNIXX-----A---8', 'A2--------A----', 'AAXXX----1A---8', 'C}-------------', 'NNFXX-----A---8', 'Db------------8'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----', 'Vi-P---2--A----', 'Vi-P---1--A----', 'NNIPX-----A----', 'NNFSX-----A----', 'AAXXX----1A----', 'Xx-------------']
474: ['AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------', 'NNMS5-----A----', 'NNFPX-----A----', 'NNFS1-----A---1', 'AUFS1M---------', 'RR--1----------', 'RR--X----------', 'XX-------------', 'X@-------------'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----', 'Vi-P---2--A----', 'Vi-P---1--A----', 'NNIPX-----A----', 'NNFSX-----A----', 'AAXXX----1A----', 'Xx-------------', 'NNMXX-----A---8', 'NNIXX-----A---8', 'A2--------A----', 'AAXXX----1A---8', 'C}-------------', 'NNFXX-----A---8', 'Db------------8']
475: ['VB-S---3P-AA---'] into ['J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---']
476: ['VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---', 'VB-P---1P-AA---'] into ['VB-S---1P-AA---', 'Vc-S---1-------', 'Vc-X---3-------']
477: ['PSXXXZS3-------'] into ['P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1']
478: ['PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----'] into ['AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----']
479: ['J,-------------'] into ['P4XP3----------', 'PJXP1----------', 'P4NS1----------', 'PQ--1----------', 'J,-X---3-------', 'P4IP1----------', 'P4FP1----------', 'P4MP1----------']
480: ['NNNS4-----A----'] into ['AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----']
481: ['PKM-3----------', 'PQ--2----------', 'P4ZS7----------', 'PKM-1----------', 'PQ--4----------', 'Vi-P---2--N----', 'J,-P---1-------', 'PJYS2----------', 'P4MS4----------', 'PJZS3----------', 'P4YP4----------', 'P4FS4----------', 'P4FP4----------', 'P4IS4----------', 'J,-S---1-------', 'P4NS4----------', 'PJYS1----------', 'PJNS1----------', 'P4YS1----------', 'PE--1----------', 'P4FS1----------'] into ['P4XP3----------', 'PJXP1----------', 'P4NS1----------', 'PQ--1----------', 'J,-X---3-------', 'P4IP1----------', 'P4FP1----------', 'P4MP1----------', 'J,-------------']
482: ['AAIP2----1N----', 'AAIP2----2A----', 'AGIP2-----A----', 'AAIP2----1A----'] into ['AGMP2-----A----', 'AAMP2----1N----', 'AAMP2----1A----']
483: ['NNIP2-----A----'] into ['NNMP2-----A----', 'NNNP2-----A----']
484: ['AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------', 'CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------', 'AGFP2-----A----', 'AAFP2----1N----', 'PDXP2----------', 'AANP2----1A----', 'PZXP2----------', 'P8XP2----------', 'Ca--2----------', 'AAMP2----3A----', 'PLXP2----------', 'ClXS2----------', 'Ca--4----------', 'Ca--1----------', 'Cn-S1----------', 'Cn-S4----------'] into ['AGMP2-----A----', 'AAMP2----1N----', 'AAMP2----1A----', 'AAIP2----1N----', 'AAIP2----2A----', 'AGIP2-----A----', 'AAIP2----1A----']
485: ['NNNS6-----A----'] into ['NNIS6-----A---1', 'NNIS6-----A----']
486: ['PSZS6FS3-------', 'AAIS6----1N----', 'ClZS6----------', 'CrIS6----------', 'PSZS6-P1-------', 'AUNS6M---------', 'CrNS6----------', 'AANS6----1A----'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----', 'P9FS6----------', 'P5ZS6--3-------', 'PQ--6----------', 'P9XP6----------', 'P9ZS6----------', 'PDZS6----------', 'AAIS6----1A----']
487: ['Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----'] into ['AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------']
488: ['NNFS7-----A----'] into ['NNIS7-----A----']
489: ['NNMS1-----A----'] into ['AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----', 'PSYS1-P1-------', 'PSYS1-S1-------', 'AAMS1----1A----']
490: ['AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----', 'PSYS1-P1-------', 'PSYS1-S1-------', 'AAMS1----1A----', 'NNMS1-----A----'] into ['AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----', 'VeYS------A----', 'CrIS7----------', 'PDZS7----------', 'AAIS7----1A----']
491: ['AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----', 'VeYS------A----', 'CrIS7----------', 'PDZS7----------', 'AAIS7----1A----', 'AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----', 'PSYS1-P1-------', 'PSYS1-S1-------', 'AAMS1----1A----', 'NNMS1-----A----'] into ['P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----', 'Cn-P7----------', 'P8XP7----------', 'ClXP7----------', 'AAFP7----2A----', 'AAFP7----1A----', 'AAIP7----1A----', 'AAMS7----1A----', 'AAMS7----3A----', 'AANS7----1N----', 'AANS7----1A----']
492: ['P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----', 'Cn-P7----------', 'P8XP7----------', 'ClXP7----------', 'AAFP7----2A----', 'AAFP7----1A----', 'AAIP7----1A----', 'AAMS7----1A----', 'AAMS7----3A----', 'AANS7----1N----', 'AANS7----1A----', 'AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----', 'VeYS------A----', 'CrIS7----------', 'PDZS7----------', 'AAIS7----1A----', 'AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----', 'PSYS1-P1-------', 'PSYS1-S1-------', 'AAMS1----1A----', 'NNMS1-----A----'] into ['NNIP7-----A----', 'AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----', 'CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----', 'VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----', 'AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----', 'P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----', 'P5ZS7--3-------', 'P5XP7--3-------', 'NNMP7-----A----', 'NNFP7-----A----', 'NNMS7-----A----', 'NNNS7-----A----']
493: ['NNIXX-----A----', 'NNNXX-----A----', 'NNFXX-----A----'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----', 'Vi-P---2--A----', 'Vi-P---1--A----', 'NNIPX-----A----', 'NNFSX-----A----', 'AAXXX----1A----', 'Xx-------------', 'NNMXX-----A---8', 'NNIXX-----A---8', 'A2--------A----', 'AAXXX----1A---8', 'C}-------------', 'NNFXX-----A---8', 'Db------------8', 'AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------', 'NNMS5-----A----', 'NNFPX-----A----', 'NNFS1-----A---1', 'AUFS1M---------', 'RR--1----------', 'RR--X----------', 'XX-------------', 'X@-------------']
494: ['AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----'] into ['AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----']
495: ['NNIP3-----A----', 'NNFP3-----A----', 'NNMP3-----A----', 'NNFS3-----A----'] into ['AANS3----1A----', 'NNNS3-----A----', 'AAIS3----1A----', 'PDZS3----------', 'NNIS3-----A----']
496: ['PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----'] into ['ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----']
497: ['PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------'] into ['PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------']
498: ['VpYS---XR-AA---', 'VpYS---XR-AA--1'] into ['VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---']
499: ['NNIP1-----A----'] into ['AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----']
500: ['AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----'] into ['AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----']
501: ['AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----'] into ['P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------']
502: ['VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---'] into ['VpQW---XR-AA---', 'VpNS---XR-AA---']
503: ['AUNS1M---------', 'CrNS1----------', 'AANS1----1A----'] into ['VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---']
504: ['NNNS1-----A----'] into ['NNMP1-----A---1', 'NNMP1-----A----']
505: ['J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---', 'VB-S---3P-AA---'] into ['VpQW---XR-AA---', 'VpNS---XR-AA---', 'VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---']
506: ['NNIS2-----A----'] into ['NNNS2-----A----']
507: ['NNFP2-----A----'] into ['NNMP2-----A----', 'NNNP2-----A----', 'NNIP2-----A----']
508: ['AAFP2----1A----'] into ['AGMP2-----A----', 'AAMP2----1N----', 'AAMP2----1A----', 'AAIP2----1N----', 'AAIP2----2A----', 'AGIP2-----A----', 'AAIP2----1A----', 'AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------', 'CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------', 'AGFP2-----A----', 'AAFP2----1N----', 'PDXP2----------', 'AANP2----1A----', 'PZXP2----------', 'P8XP2----------', 'Ca--2----------', 'AAMP2----3A----', 'PLXP2----------', 'ClXS2----------', 'Ca--4----------', 'Ca--1----------', 'Cn-S1----------', 'Cn-S4----------']
509: ['AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------', 'Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----'] into ['PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------', 'PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------']
510: ['NNIS4-----A----'] into ['AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----', 'NNNS4-----A----']
511: ['CrIS4----------', 'ClIS4----------', 'PDIS4----------', 'AANP1----1A----', 'CrNP1----------', 'AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------', 'AAIS4----1A----'] into ['AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----', 'NNNS4-----A----', 'NNIS4-----A----']
512: ['ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----'] into ['VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----']
513: ['NNMP6-----A----', 'NNFP1-----N----', 'NNMS6-----A---1', 'AAMS6----1A----', 'NNMS6-----A----', 'NNIP6-----A----', 'NNNP6-----A----', 'NNFP6-----A----'] into ['NNIS6-----A---1', 'NNIS6-----A----', 'NNNS6-----A----']
514: ['VB-S---1P-AA---', 'Vc-S---1-------', 'Vc-X---3-------', 'VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---', 'VB-P---1P-AA---'] into ['PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------', 'PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------', 'AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------', 'Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----']
515: ['NNFP1-----A----', 'NNNP1-----A----'] into ['NNMP1-----A---1', 'NNMP1-----A----', 'NNNS1-----A----']
516: ['AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------', 'AAFP1----1A----'] into ['AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----', 'AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----']
517: ['NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----'] into ['ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----']
518: ['ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----'] into ['P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----']
519: ['P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----'] into ['NNNS2-----A----', 'NNIS2-----A----']
520: ['NNNS2-----A----', 'NNIS2-----A----', 'P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----'] into ['VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----', 'ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----']
521: ['VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----', 'ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----', 'NNNS2-----A----', 'NNIS2-----A----', 'P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----'] into ['AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----', 'AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----', 'AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------', 'AAFP1----1A----']
522: ['NNIS7-----A----', 'NNFS7-----A----'] into ['NNIP7-----A----', 'AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----', 'CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----', 'VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----', 'AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----', 'P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----', 'P5ZS7--3-------', 'P5XP7--3-------', 'NNMP7-----A----', 'NNFP7-----A----', 'NNMS7-----A----', 'NNNS7-----A----', 'P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----', 'Cn-P7----------', 'P8XP7----------', 'ClXP7----------', 'AAFP7----2A----', 'AAFP7----1A----', 'AAIP7----1A----', 'AAMS7----1A----', 'AAMS7----3A----', 'AANS7----1N----', 'AANS7----1A----', 'AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----', 'VeYS------A----', 'CrIS7----------', 'PDZS7----------', 'AAIS7----1A----', 'AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----', 'PSYS1-P1-------', 'PSYS1-S1-------', 'AAMS1----1A----', 'NNMS1-----A----']
523: ['PSFS6-P1-------', 'AAFS6----1N----', 'CrFS6----------', 'AUFS6M---------', 'AAFS6----1A----'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----', 'P9FS6----------', 'P5ZS6--3-------', 'PQ--6----------', 'P9XP6----------', 'P9ZS6----------', 'PDZS6----------', 'AAIS6----1A----', 'PSZS6FS3-------', 'AAIS6----1N----', 'ClZS6----------', 'CrIS6----------', 'PSZS6-P1-------', 'AUNS6M---------', 'CrNS6----------', 'AANS6----1A----']
524: ['RR--3----------', 'RV--3----------'] into ['RR--7----------', 'RV--7----------']
525: ['AANS3----1A----', 'NNNS3-----A----', 'AAIS3----1A----', 'PDZS3----------', 'NNIS3-----A----', 'NNIP3-----A----', 'NNFP3-----A----', 'NNMP3-----A----', 'NNFS3-----A----'] into ['NNIP7-----A----', 'AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----', 'CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----', 'VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----', 'AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----', 'P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----', 'P5ZS7--3-------', 'P5XP7--3-------', 'NNMP7-----A----', 'NNFP7-----A----', 'NNMS7-----A----', 'NNNS7-----A----', 'P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----', 'Cn-P7----------', 'P8XP7----------', 'ClXP7----------', 'AAFP7----2A----', 'AAFP7----1A----', 'AAIP7----1A----', 'AAMS7----1A----', 'AAMS7----3A----', 'AANS7----1N----', 'AANS7----1A----', 'AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----', 'VeYS------A----', 'CrIS7----------', 'PDZS7----------', 'AAIS7----1A----', 'AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----', 'PSYS1-P1-------', 'PSYS1-S1-------', 'AAMS1----1A----', 'NNMS1-----A----', 'NNIS7-----A----', 'NNFS7-----A----']
526: ['NNFS1-----A----'] into ['NNIS1-----A----']
527: ['PDYS1----------', 'PLYS1----------', 'PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------', 'CrIS1----------', 'AAIS1----3A----', 'AAIS1----1A----'] into ['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------', 'PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------', 'AANS1----2A----', 'AANS1----1N----', 'Vf--------N----', 'AAFS1----1N----', 'AAFS1----2A----', 'AAFS1----1A----']
528: ['C=-------------'] into ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----', 'Vi-P---2--A----', 'Vi-P---1--A----', 'NNIPX-----A----', 'NNFSX-----A----', 'AAXXX----1A----', 'Xx-------------', 'NNMXX-----A---8', 'NNIXX-----A---8', 'A2--------A----', 'AAXXX----1A---8', 'C}-------------', 'NNFXX-----A---8', 'Db------------8', 'AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------', 'NNMS5-----A----', 'NNFPX-----A----', 'NNFS1-----A---1', 'AUFS1M---------', 'RR--1----------', 'RR--X----------', 'XX-------------', 'X@-------------', 'NNIXX-----A----', 'NNNXX-----A----', 'NNFXX-----A----']
529: ['RV--4----------', 'RR--4----------'] into ['AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------', 'PDFS4----------', 'AAFS4----2A----', 'ClFS4----------', 'P8FS4----------', 'AAFS4----1N----', 'PZFS4----------', 'PWFS4----------', 'PSFS4-P1-------', 'P8FS4---------1', 'AAFS4----1A----']
530: ['NNMP1-----A---1', 'NNMP1-----A----', 'NNNS1-----A----', 'NNFP1-----A----', 'NNNP1-----A----'] into ['NNIS1-----A----', 'NNFS1-----A----']
531: ['NNFS2-----A----'] into ['AUIS2M---------', 'AAIS2----1N----', 'AAIS2----1A----']
532: ['J^-------------'] into ['P7-X3----------', 'P7-X4----------']
533: ['VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---', 'VpYS---XR-AA---', 'VpYS---XR-AA--1'] into ['P7-X3----------', 'P7-X4----------', 'J^-------------']
534: ['VpQW---XR-AA---', 'VpNS---XR-AA---', 'VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---', 'J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---', 'VB-S---3P-AA---'] into ['P7-X3----------', 'P7-X4----------', 'J^-------------', 'VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---', 'VpYS---XR-AA---', 'VpYS---XR-AA--1']
535: ['NNFS6-----A----'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----', 'P9FS6----------', 'P5ZS6--3-------', 'PQ--6----------', 'P9XP6----------', 'P9ZS6----------', 'PDZS6----------', 'AAIS6----1A----', 'PSZS6FS3-------', 'AAIS6----1N----', 'ClZS6----------', 'CrIS6----------', 'PSZS6-P1-------', 'AUNS6M---------', 'CrNS6----------', 'AANS6----1A----', 'PSFS6-P1-------', 'AAFS6----1N----', 'CrFS6----------', 'AUFS6M---------', 'AAFS6----1A----']
536: ['NNIS6-----A---1', 'NNIS6-----A----', 'NNNS6-----A----', 'NNMP6-----A----', 'NNFP1-----N----', 'NNMS6-----A---1', 'AAMS6----1A----', 'NNMS6-----A----', 'NNIP6-----A----', 'NNNP6-----A----', 'NNFP6-----A----'] into ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----', 'P9FS6----------', 'P5ZS6--3-------', 'PQ--6----------', 'P9XP6----------', 'P9ZS6----------', 'PDZS6----------', 'AAIS6----1A----', 'PSZS6FS3-------', 'AAIS6----1N----', 'ClZS6----------', 'CrIS6----------', 'PSZS6-P1-------', 'AUNS6M---------', 'CrNS6----------', 'AANS6----1A----', 'PSFS6-P1-------', 'AAFS6----1N----', 'CrFS6----------', 'AUFS6M---------', 'AAFS6----1A----', 'NNFS6-----A----']
537: ['NNFS4-----A----'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----', 'PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----', 'AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----', 'Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----']
538: ['AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----', 'NNNS4-----A----', 'NNIS4-----A----', 'CrIS4----------', 'ClIS4----------', 'PDIS4----------', 'AANP1----1A----', 'CrNP1----------', 'AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------', 'AAIS4----1A----'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----', 'PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----', 'AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----', 'Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----', 'NNFS4-----A----']
539: ['PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------', 'PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------', 'AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------', 'Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----', 'VB-S---1P-AA---', 'Vc-S---1-------', 'Vc-X---3-------', 'VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---', 'VB-P---1P-AA---'] into ['P7-X3----------', 'P7-X4----------', 'J^-------------', 'VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---', 'VpYS---XR-AA---', 'VpYS---XR-AA--1', 'VpQW---XR-AA---', 'VpNS---XR-AA---', 'VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---', 'J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---', 'VB-S---3P-AA---']
540: ['P7-X3----------', 'P7-X4----------', 'J^-------------', 'VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---', 'VpYS---XR-AA---', 'VpYS---XR-AA--1', 'VpQW---XR-AA---', 'VpNS---XR-AA---', 'VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---', 'J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---', 'VB-S---3P-AA---', 'PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------', 'PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------', 'AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------', 'Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----', 'VB-S---1P-AA---', 'Vc-S---1-------', 'Vc-X---3-------', 'VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---', 'VB-P---1P-AA---'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----', 'PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----', 'AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----', 'Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----', 'NNFS4-----A----', 'AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----', 'NNNS4-----A----', 'NNIS4-----A----', 'CrIS4----------', 'ClIS4----------', 'PDIS4----------', 'AANP1----1A----', 'CrNP1----------', 'AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------', 'AAIS4----1A----']
541: ['AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------', 'PDFS4----------', 'AAFS4----2A----', 'ClFS4----------', 'P8FS4----------', 'AAFS4----1N----', 'PZFS4----------', 'PWFS4----------', 'PSFS4-P1-------', 'P8FS4---------1', 'AAFS4----1A----', 'RV--4----------', 'RR--4----------'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----', 'PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----', 'AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----', 'Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----', 'NNFS4-----A----', 'AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----', 'NNNS4-----A----', 'NNIS4-----A----', 'CrIS4----------', 'ClIS4----------', 'PDIS4----------', 'AANP1----1A----', 'CrNP1----------', 'AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------', 'AAIS4----1A----', 'P7-X3----------', 'P7-X4----------', 'J^-------------', 'VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---', 'VpYS---XR-AA---', 'VpYS---XR-AA--1', 'VpQW---XR-AA---', 'VpNS---XR-AA---', 'VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---', 'J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---', 'VB-S---3P-AA---', 'PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------', 'PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------', 'AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------', 'Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----', 'VB-S---1P-AA---', 'Vc-S---1-------', 'Vc-X---3-------', 'VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---', 'VB-P---1P-AA---']
542: ['AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----', 'AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----', 'AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------', 'AAFP1----1A----', 'VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----', 'ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----', 'NNNS2-----A----', 'NNIS2-----A----', 'P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----'] into ['AAFS2----1A----']
543: ['AAFS2----1A----', 'AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----', 'AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----', 'AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------', 'AAFP1----1A----', 'VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----', 'ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----', 'NNNS2-----A----', 'NNIS2-----A----', 'P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----'] into ['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------', 'PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------', 'AANS1----2A----', 'AANS1----1N----', 'Vf--------N----', 'AAFS1----1N----', 'AAFS1----2A----', 'AAFS1----1A----', 'PDYS1----------', 'PLYS1----------', 'PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------', 'CrIS1----------', 'AAIS1----3A----', 'AAIS1----1A----']
544: ['AUIS2M---------', 'AAIS2----1N----', 'AAIS2----1A----', 'NNFS2-----A----'] into ['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------', 'PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------', 'AANS1----2A----', 'AANS1----1N----', 'Vf--------N----', 'AAFS1----1N----', 'AAFS1----2A----', 'AAFS1----1A----', 'PDYS1----------', 'PLYS1----------', 'PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------', 'CrIS1----------', 'AAIS1----3A----', 'AAIS1----1A----', 'AAFS2----1A----', 'AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----', 'AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----', 'AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------', 'AAFP1----1A----', 'VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----', 'ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----', 'NNNS2-----A----', 'NNIS2-----A----', 'P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----']
545: ['AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------', 'PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------', 'AANS1----2A----', 'AANS1----1N----', 'Vf--------N----', 'AAFS1----1N----', 'AAFS1----2A----', 'AAFS1----1A----', 'PDYS1----------', 'PLYS1----------', 'PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------', 'CrIS1----------', 'AAIS1----3A----', 'AAIS1----1A----', 'AAFS2----1A----', 'AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----', 'AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----', 'AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------', 'AAFP1----1A----', 'VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----', 'ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----', 'NNNS2-----A----', 'NNIS2-----A----', 'P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----', 'AUIS2M---------', 'AAIS2----1N----', 'AAIS2----1A----', 'NNFS2-----A----'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----', 'PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----', 'AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----', 'Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----', 'NNFS4-----A----', 'AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----', 'NNNS4-----A----', 'NNIS4-----A----', 'CrIS4----------', 'ClIS4----------', 'PDIS4----------', 'AANP1----1A----', 'CrNP1----------', 'AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------', 'AAIS4----1A----', 'P7-X3----------', 'P7-X4----------', 'J^-------------', 'VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---', 'VpYS---XR-AA---', 'VpYS---XR-AA--1', 'VpQW---XR-AA---', 'VpNS---XR-AA---', 'VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---', 'J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---', 'VB-S---3P-AA---', 'PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------', 'PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------', 'AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------', 'Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----', 'VB-S---1P-AA---', 'Vc-S---1-------', 'Vc-X---3-------', 'VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---', 'VB-P---1P-AA---', 'AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------', 'PDFS4----------', 'AAFS4----2A----', 'ClFS4----------', 'P8FS4----------', 'AAFS4----1N----', 'PZFS4----------', 'PWFS4----------', 'PSFS4-P1-------', 'P8FS4---------1', 'AAFS4----1A----', 'RV--4----------', 'RR--4----------']
546: ['AAFS2----2A----', 'CrFS2----------', 'AUFS2M---------', 'RV--2----------', 'RR--2----------'] into ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----', 'PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----', 'AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----', 'Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----', 'NNFS4-----A----', 'AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----', 'NNNS4-----A----', 'NNIS4-----A----', 'CrIS4----------', 'ClIS4----------', 'PDIS4----------', 'AANP1----1A----', 'CrNP1----------', 'AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------', 'AAIS4----1A----', 'P7-X3----------', 'P7-X4----------', 'J^-------------', 'VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---', 'VpYS---XR-AA---', 'VpYS---XR-AA--1', 'VpQW---XR-AA---', 'VpNS---XR-AA---', 'VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---', 'J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---', 'VB-S---3P-AA---', 'PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------', 'PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------', 'AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------', 'Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----', 'VB-S---1P-AA---', 'Vc-S---1-------', 'Vc-X---3-------', 'VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---', 'VB-P---1P-AA---', 'AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------', 'PDFS4----------', 'AAFS4----2A----', 'ClFS4----------', 'P8FS4----------', 'AAFS4----1N----', 'PZFS4----------', 'PWFS4----------', 'PSFS4-P1-------', 'P8FS4---------1', 'AAFS4----1A----', 'RV--4----------', 'RR--4----------', 'AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------', 'PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------', 'AANS1----2A----', 'AANS1----1N----', 'Vf--------N----', 'AAFS1----1N----', 'AAFS1----2A----', 'AAFS1----1A----', 'PDYS1----------', 'PLYS1----------', 'PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------', 'CrIS1----------', 'AAIS1----3A----', 'AAIS1----1A----', 'AAFS2----1A----', 'AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----', 'AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----', 'AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------', 'AAFP1----1A----', 'VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----', 'ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----', 'NNNS2-----A----', 'NNIS2-----A----', 'P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----', 'AUIS2M---------', 'AAIS2----1N----', 'AAIS2----1A----', 'NNFS2-----A----']
547: ['NNMP2-----A----', 'NNNP2-----A----', 'NNIP2-----A----', 'NNFP2-----A----'] into ['AGMP2-----A----', 'AAMP2----1N----', 'AAMP2----1A----', 'AAIP2----1N----', 'AAIP2----2A----', 'AGIP2-----A----', 'AAIP2----1A----', 'AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------', 'CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------', 'AGFP2-----A----', 'AAFP2----1N----', 'PDXP2----------', 'AANP2----1A----', 'PZXP2----------', 'P8XP2----------', 'Ca--2----------', 'AAMP2----3A----', 'PLXP2----------', 'ClXS2----------', 'Ca--4----------', 'Ca--1----------', 'Cn-S1----------', 'Cn-S4----------', 'AAFP2----1A----']
Class 1: ['P5XP4--3-------', 'P5FS4--3-------', 'P5ZS4--3-------', 'P6-X4----------', 'NNNP4-----A----', 'PLNS4---------1', 'PZZS7----------', 'PDMP4----------', 'PSXP4-P1-------', 'AAFP4----1N----', 'AANP4----1A----', 'AGFP6-----A----', 'PLXP6----------', 'P8XP6----------', 'AAFP6----1A----', 'PSXP6-P1-------', 'Cn-P6----------', 'ClXP6----------', 'AANP6----1A----', 'Ca--6----------', 'PZXP6----------', 'PDXP6----------', 'AAIP6----1A----', 'AAIP6----3A----', 'PDFS6----------', 'AAFS6----3A----', 'P8ZS6----------', 'P4ZS6----------', 'AUIS6M---------', 'PLZS6----------', 'P4FS6----------', 'AGFS6-----A----', 'P8FS6---------1', 'ClFS6----------', 'PZFS6----------', 'AGIS6-----A----', 'P1XXXZS3-------', 'P1XXXXP3-------', 'PSXXXXP3-------', 'PDFS3----------', 'AAFS3----1A----', 'AAMS2----1A----', 'NNMS2-----A----', 'NNFS4-----A----', 'AANS4----1N----', 'CrNS4----------', 'P8NS4---------1', 'AANS4----1A----', 'P8MS4----------', 'ClMS4----------', 'AAMS4----1A----', 'NNMS4-----A----', 'NNNS4-----A----', 'NNIS4-----A----', 'CrIS4----------', 'ClIS4----------', 'PDIS4----------', 'AANP1----1A----', 'CrNP1----------', 'AAIS4----3A----', 'AGIS4-----A----', 'P8IS4----------', 'AAIS4----1N----', 'PZIS4----------', 'AAIS4----1A----', 'P7-X3----------', 'P7-X4----------', 'J^-------------', 'VpMP---XR-NA---', 'VpMP---XR-AA--1', 'VB-P---3P-NA---', 'VpMP---XR-AA---', 'VpYS---XR-AA---', 'VpYS---XR-AA--1', 'VpQW---XR-AA---', 'VpNS---XR-AA---', 'VB-P---3P-NA--1', 'VB-P---3F-NA---', 'VpTP---XR-NA---', 'VpNS---XR-AA--1', 'VpTP---XR-AA--1', 'VpTP---XR-AA---', 'VB-P---3P-AA---', 'J^------------1', 'PQ--7----------', 'VB-S---3F-NA---', 'NNIS5-----A----', 'NNFP2-----A---1', 'AANP7----1A----', 'VpNS---XR-NA---', 'VB-S---1F-AA---', 'ACQW------A----', 'PZM-4----------', 'VB-P---1F-AA---', 'VB-S---1P-NA--1', 'VB-P---3P-AA--1', 'VB-P---3F-AA---', 'VpYS---XR-NA---', 'VB-S---3P-NA---', 'VpQW---XR-AA--1', 'VpQW---XR-NA---', 'VB-S---3F-AA---', 'VB-S---3P-AA---', 'PLNS1----------', 'PWM-1----------', 'PW--4----------', 'PW--1----------', 'ClNS1----------', 'PLNS1---------1', 'NNIS2-----A---1', 'P6-X2----------', 'P4XP2----------', 'P5ZS2--3-------', 'P5XP2--3-------', 'PP-P2--1-------', 'AGIP1-----A----', 'PPYS1--3-------', 'NNMP2-----A---1', 'PLXP7----------', 'NNNS1-----N----', 'AGNS2-----A----', 'C?--4----------', 'RF-------------', 'AAIP1----2A----', 'PP-P1--2-------', 'PP-P1--1-------', 'PP-S1--1-------', 'PDNS1----------', 'PPZS4--3------2', 'PPXP4--3-------', 'P8ZS3----------', 'PPNS4--3-------', 'PPZS7--3-------', 'PPXP2--3-------', 'PHZS4--3-------', 'PPFS4--3-------', 'PP-P3--2-------', 'PZM-1----------', 'PPXP3--3-------', 'PH-S3--1-------', 'PP-P3--1-------', 'PZ--1----------', 'PPFS3--3-------', 'PHZS3--3-------', 'PP-S4--1-------', 'PP-P4--2-------', 'PP-P4--1-------', 'PDNS4----------', 'AGNS1-----A----', 'VB-S---2P-AA---', 'PLNS4----------', 'AANS1----3A----', 'ACNS------A----', 'ClFS7----------', 'ClZS7----------', 'Dg-------3A----', 'TT-------------', 'Cv-------------', 'Db------------1', 'Db-------------', 'Dg-------1A---3', 'Co-------------', 'Dg-------1A---1', 'Dg-------3A---1', 'PH-S4--1-------', 'Dg-------2A---1', 'Dg-------2A----', 'ACYS------A----', 'PZ--4----------', 'Dg-------1N----', 'Dg-------1A----', 'VB-S---1P-AA---', 'Vc-S---1-------', 'Vc-X---3-------', 'VB-S---1P-NA---', 'VB-P---2P-AA---', 'VB-S---1P-AA--1', 'VB-P---1P-NA---', 'VB-P---1P-AA---', 'AAFS4----3A----', 'AUFS4M---------', 'CrFS4----------', 'PDFS4----------', 'AAFS4----2A----', 'ClFS4----------', 'P8FS4----------', 'AAFS4----1N----', 'PZFS4----------', 'PWFS4----------', 'PSFS4-P1-------', 'P8FS4---------1', 'AAFS4----1A----', 'RV--4----------', 'RR--4----------', 'AAFS1----3A----', 'ClFS1----------', 'AGFS1-----A----', 'PSFSXFS3-------', 'CrFS1----------', 'PWFS1----------', 'PZFS1----------', 'PSHS1-P1-------', 'PSHS1-S1-------', 'PDFS1----------', 'AANS1----2A----', 'AANS1----1N----', 'Vf--------N----', 'AAFS1----1N----', 'AAFS1----2A----', 'AAFS1----1A----', 'PDYS1----------', 'PLYS1----------', 'PZYS1----------', 'PSZS1FS3-------', 'PWYS1----------', 'AUIS1M---------', 'CrIS1----------', 'AAIS1----3A----', 'AAIS1----1A----', 'AAFS2----1A----', 'AANP3----1A----', 'P8ZS7----------', 'AAIP3----1A----', 'P6-X3----------', 'ACTP------A----', 'PLFS1----------', 'P5ZS3--3-------', 'P5FS3--3-------', 'ClXP3----------', 'PLXP3----------', 'PDXP3----------', 'P8XP3----------', 'AAMP3----1A----', 'NNFS1-----N----', 'NNIP2-----A---1', 'NNNP3-----A----', 'AAFP3----1A----', 'AAFS7----1N----', 'AUFS7M---------', 'PDFS7----------', 'AAFS7----3A----', 'AGFS7-----A----', 'P8FS7---------1', 'AAFS7----1A----', 'AAIP1----1N----', 'AAIP1----3A----', 'CrIP1----------', 'PSHP1-P2-------', 'PDIP1----------', 'PLIP1----------', 'ClYP1----------', 'AAIP1----1A----', 'PLMP1----------', 'PDMP1----------', 'AGMP1-----A----', 'AAMP1----3A----', 'PSMP1-P1-------', 'AAMP1----2A----', 'PPMP1--3-------', 'PZMP1----------', 'AAMP1----1A----', 'NNIP1-----A----', 'AAFP1----2A----', 'AAFP1----1N----', 'ClXP1----------', 'CrFP1----------', 'AGFP1-----A----', 'AAMP1----1N----', 'NNNP1-----A---2', 'AAIP7----3A----', 'Vf--------A---1', 'PLFP1----------', 'AAFP1----3A----', 'ClHP1----------', 'PSHP1-P1-------', 'PZFP1----------', 'PDFP1----------', 'AAFP1----1A----', 'VsNS---XX-AP---', 'VsQW---XX-AP---', 'AGIS1-----A----', 'ClYS1----------', 'AAIS1----2A----', 'AAIS1----1N----', 'VsYS---XX-AP---', 'AUNS1M---------', 'CrNS1----------', 'AANS1----1A----', 'ACMP------A----', 'Vc-P---1-------', 'VsMP---XX-AP---', 'VsTP---XX-AP---', 'Vf--------A----', 'NNNS2-----A----', 'NNIS2-----A----', 'P5ZS2--3------1', 'P5FS2--3-------', 'P9XP2----------', 'PDZS2----------', 'P8ZS2----------', 'PSZS2-P1-------', 'PSZS2FS3-------', 'PSZS2-S1-------', 'PSFS2-P1-------', 'AGFS2-----A----', 'PDFS2----------', 'AAFS2----1N----', 'ClFS2----------', 'P8FS2---------1', 'PSXXXZS3-------', 'AUNS2M---------', 'CrNS2----------', 'AANS2----1N----', 'AANS2----1A----', 'ClHP4----------', 'PKM-4----------', 'PDFP4----------', 'PWFP4----------', 'PZFP4----------', 'PLFP4----------', 'P8FP4---------1', 'AAFP4----1A----', 'PZYP4----------', 'AAIP4----1N----', 'PDIP4----------', 'AAIP4----1A----', 'ClXP4----------', 'ClYP4----------', 'PLYP4----------', 'P8YP4---------1', 'AAMP4----1A----', 'NNMP4-----A----', 'NNIP4-----A----', 'NNFP4-----A----', 'AUIS2M---------', 'AAIS2----1N----', 'AAIS2----1A----', 'NNFS2-----A----', 'AAFS2----2A----', 'CrFS2----------', 'AUFS2M---------', 'RV--2----------', 'RR--2----------']
Class 2: ['NNIP7-----A----', 'AGNP6-----A----', 'NNFS2-----N----', 'NNFS2-----A---1', 'NNFS4-----N----', 'CrIS2----------', 'ClZS2----------', 'AAIS2----3A----', 'AAIS2----2A----', 'VB-P---2P-NA---', 'PDMS4----------', 'ClZS3----------', 'AGIS2-----A----', 'AGFS4-----A----', 'AAMS3----1A----', 'NNMS3-----A---1', 'NNMS3-----A----', 'P6-X7----------', 'PP-P7--1-------', 'NNNP7-----A----', 'P5ZS7--3-------', 'P5XP7--3-------', 'NNMP7-----A----', 'NNFP7-----A----', 'NNMS7-----A----', 'NNNS7-----A----', 'P9ZS7----------', 'P4FS7----------', 'AGFP7-----A----', 'AAMP7----1A----', 'Cn-P7----------', 'P8XP7----------', 'ClXP7----------', 'AAFP7----2A----', 'AAFP7----1A----', 'AAIP7----1A----', 'AAMS7----1A----', 'AAMS7----3A----', 'AANS7----1N----', 'AANS7----1A----', 'AAIS7----1N----', 'PDXP7----------', 'AAIS7----3A----', 'AGIS7-----A----', 'VeYS------A----', 'CrIS7----------', 'PDZS7----------', 'AAIS7----1A----', 'AAMS1----1N----', 'AUMS1M---------', 'AGMS1-----A----', 'CrMS1----------', 'AAMS1----2A----', 'PSYS1-P1-------', 'PSYS1-S1-------', 'AAMS1----1A----', 'NNMS1-----A----', 'NNIS7-----A----', 'NNFS7-----A----', 'AANS3----1A----', 'NNNS3-----A----', 'AAIS3----1A----', 'PDZS3----------', 'NNIS3-----A----', 'NNIP3-----A----', 'NNFP3-----A----', 'NNMP3-----A----', 'NNFS3-----A----']
Class 3: ['P5FS6--3-------', 'P5XP6--3-------', 'P6-X6----------', 'NNNSX-----A----', 'P9FS6----------', 'P5ZS6--3-------', 'PQ--6----------', 'P9XP6----------', 'P9ZS6----------', 'PDZS6----------', 'AAIS6----1A----', 'PSZS6FS3-------', 'AAIS6----1N----', 'ClZS6----------', 'CrIS6----------', 'PSZS6-P1-------', 'AUNS6M---------', 'CrNS6----------', 'AANS6----1A----', 'PSFS6-P1-------', 'AAFS6----1N----', 'CrFS6----------', 'AUFS6M---------', 'AAFS6----1A----', 'NNFS6-----A----', 'NNIS6-----A---1', 'NNIS6-----A----', 'NNNS6-----A----', 'NNMP6-----A----', 'NNFP1-----N----', 'NNMS6-----A---1', 'AAMS6----1A----', 'NNMS6-----A----', 'NNIP6-----A----', 'NNNP6-----A----', 'NNFP6-----A----']
Class 4: ['AGMP2-----A----', 'AAMP2----1N----', 'AAMP2----1A----', 'AAIP2----1N----', 'AAIP2----2A----', 'AGIP2-----A----', 'AAIP2----1A----', 'AAFP2----2A----', 'AAFP2----3A----', 'PSXP2-P1-------', 'ClXP2----------', 'CrNP2----------', 'AGNP2-----A----', 'Cn-P2----------', 'AGFP2-----A----', 'AAFP2----1N----', 'PDXP2----------', 'AANP2----1A----', 'PZXP2----------', 'P8XP2----------', 'Ca--2----------', 'AAMP2----3A----', 'PLXP2----------', 'ClXS2----------', 'Ca--4----------', 'Ca--1----------', 'Cn-S1----------', 'Cn-S4----------', 'AAFP2----1A----', 'NNMP2-----A----', 'NNNP2-----A----', 'NNIP2-----A----', 'NNFP2-----A----']
Class 5: ['P4XP3----------', 'PJXP1----------', 'P4NS1----------', 'PQ--1----------', 'J,-X---3-------', 'P4IP1----------', 'P4FP1----------', 'P4MP1----------', 'J,-------------', 'PKM-3----------', 'PQ--2----------', 'P4ZS7----------', 'PKM-1----------', 'PQ--4----------', 'Vi-P---2--N----', 'J,-P---1-------', 'PJYS2----------', 'P4MS4----------', 'PJZS3----------', 'P4YP4----------', 'P4FS4----------', 'P4FP4----------', 'P4IS4----------', 'J,-S---1-------', 'P4NS4----------', 'PJYS1----------', 'PJNS1----------', 'P4YS1----------', 'PE--1----------', 'P4FS1----------']
Class 6: ['NNFS5-----A----', 'II-------------', 'VB-S---2P-NA---', 'J^------------8', 'Vi-S---2--A----', 'Vi-P---2--A----', 'Vi-P---1--A----', 'NNIPX-----A----', 'NNFSX-----A----', 'AAXXX----1A----', 'Xx-------------', 'NNMXX-----A---8', 'NNIXX-----A---8', 'A2--------A----', 'AAXXX----1A---8', 'C}-------------', 'NNFXX-----A---8', 'Db------------8', 'AAIS1----1A---6', 'NNISX-----A----', 'CyFP1----------', 'NNMS5-----A----', 'NNFPX-----A----', 'NNFS1-----A---1', 'AUFS1M---------', 'RR--1----------', 'RR--X----------', 'XX-------------', 'X@-------------', 'NNIXX-----A----', 'NNNXX-----A----', 'NNFXX-----A----', 'C=-------------']
Class 7: ['Z:-------------']
Class 8: ['NNIS1-----A----', 'NNFS1-----A----', 'NNMP1-----A---1', 'NNMP1-----A----', 'NNNS1-----A----', 'NNFP1-----A----', 'NNNP1-----A----']
Class 9: ['RR--7----------', 'RV--7----------', 'RR--3----------', 'RV--3----------']
Class 10: ['RR--6----------', 'RV--6----------']


In [ ]:
for mode in ["words", "pos"]:
    if mode == "words":
        LIMIT = 1
        MIN_FREQ = 10
        K = 8000
    elif mode == "pos":
        LIMIT = 1
        MIN_FREQ = 5
        K = 200000

    for lang in ["en", "cz"]:  

        print(f"""Language: {lang}.
Merging {mode} into {LIMIT} classes.
Considering first {K} words with min. freq. {MIN_FREQ}...\n""")
        
        classes, history_of_merges = execute_merging(lang, mode, LIMIT, MIN_FREQ, K, verbose=True)

        for i, cls in enumerate(classes):
            print(f"Class {i+1}: {cls}")
        print("\n")

        ## save history to a file
        with open(f"../results/{lang}_{mode}_{LIMIT}_history.txt", "w", encoding="utf-8") as f:
            for cls_one, cls_two in history_of_merges:
                f.write(f"{cls_two} into {cls_one}\n")

Interpretation¶

Merging words¶

The merging algorithm is having (optically) better results in english compared to czech. By 'better' we mean that words with similar properties are being grouped together more likely then in czech. In english, for example, even the 50th merge was:

50: ['more'] into ['even', 'less', 'much'], grouping words that clearly describe quantity.

Whereas in czech, already the 35th merge was:

35: ['státu', 'J', 'ČSFR'] into ['mezi', 'už', 'bylo', 'nás'], where it is harder to see their common property.

This applies to all later merges. Reason for that might be the following counts of the source texts:

Text File Words Unique Words
CZ.txt 222,413 42,827
EN.txt 221,099 9,608

While having roughly the same length, there are ~4 times more unique words in czech relatively to english. This diversification and (on average) globally lower bigram counts makes it more difficult for the merging algorithm to (find and) merge 'similar' words into one class. We could experiment and try to improve the merging by introducing lemmas and merging lemmas of the words instead of the exact words, or conversaly keep merging words, but introduce more than just first 8000 words in czech. We have therefore tried to run the algorithm on 4x longer text to hopefully get 'optically' better results.

In [16]:
mode = "words"
lang = "cz"
LIMIT = 15
MIN_FREQ = 10
K = 32000

print(f"""Language: {lang}.
Merging {mode} into {LIMIT} classes.
Considering first {K} words with min. freq. {MIN_FREQ}...\n""")

classes, history_of_merges = execute_merging(lang, mode, LIMIT, MIN_FREQ, K, verbose=True)

for i, cls in enumerate(classes):
    print(f"Class {i+1}: {cls}")
print("\n")

## save history to a file
with open(f"../results/{lang}_{mode}_{LIMIT}_history.txt", "w", encoding="utf-8") as f:
    for cls_one, cls_two in history_of_merges:
        f.write(f"{cls_two} into {cls_one}\n")
Language: cz.
Merging words into 15 classes.
Considering first 32000 words with min. freq. 10...

Merging language: cz, mode: words
Sum of the Q: 7.293523347895157
['C'] into ['I'], loss: 0.0005384192482320643
['Ale'] into ['Je'], loss: 0.000721102553055495
['To'] into ['Jak'], loss: 0.0007409856393401377
['J'] into ['P'], loss: 0.0007795989272068385
['OKD'] into ['kolo'], loss: 0.0008028958453116571
['protože'] into ['kterou'], loss: 0.000905530147429931
['pokud'] into ['místo'], loss: 0.0009104544614628209
['ES'] into ['KSČ'], loss: 0.0009117620858367708
['otázku'] into ['druhé'], loss: 0.000913909765557391
['Moskva'] into ['Sparta'], loss: 0.0009226531966366703
['parlamentu'] into ['podniků'], loss: 0.0009292292966721748
['toho'] into ['těch'], loss: 0.0009341058050920184
['18'] into ['1991'], loss: 0.0009366370680159043
['třeba'] into ['později'], loss: 0.0009414902759546625
['Z'] into ['Když'], loss: 0.0009535448166031399
['měla'] into ['mohl'], loss: 0.0009698336655218874
['dolarů'] into ['korun'], loss: 0.0009729264261012192
['přece'] into ['prý'], loss: 0.000982532789164596
['informace'] into ['smlouvy'], loss: 0.001026711030383539
['8'] into ['9'], loss: 0.00102930829567095
['sebe'] into ['včerejší'], loss: 0.0010344813452425085
['často'] into ['dokonce'], loss: 0.0010357889696164559
['25'] into ['programu'], loss: 0.0010389097655573894
['organizace'] into ['války'], loss: 0.0010389097655573928
['lidí'] into ['prostředí'], loss: 0.0010433475074317616
['případě'] into ['roce'], loss: 0.0010528417118531343
['všechno'] into ['teď'], loss: 0.0010756718513941638
['jeden'] into ['poslední'], loss: 0.0010778195311147848
['nich'] into ['české'], loss: 0.0010828218085466024
['IFS'] into ['nám'], loss: 0.0010900805103972733
['den'] into ['SNR'], loss: 0.001093139062229566
['tuto'] into ['této'], loss: 0.0010934109707946803
['tato'] into ['její'], loss: 0.0011014097655573894
['armády'] into ['stavu'], loss: 0.0011014097655573912
['Za'] into ["'"], loss: 0.0011085208374529889
['svých'] into ['dvou'], loss: 0.001116729296672175
['České'] into ['celé'], loss: 0.0011321110718955987
['zájem'] into ['&percnt;'], loss: 0.0011403195311147831
['12'] into ['10'], loss: 0.0011432075490481527
['proto'] into ['lidé'], loss: 0.0011458270958998734
['dnech'] into ['poslanců'], loss: 0.001146442554721988
['miliónů'] into ['tisíc'], loss: 0.0011575205437204562
['K'] into ['Kčs'], loss: 0.0011621282112493573
['tomto'] into ['úterý'], loss: 0.0011634676065215357
['majetku'] into ['ČNR'], loss: 0.0011761085007312398
['O'] into ['Po'], loss: 0.001177329636761825
['která'] into ['kde'], loss: 0.001179353129567011
['mít'] into ['některé'], loss: 0.00118535232027938
['chce'] into ['ceny'], loss: 0.00118535232027938
['může'] into ['sám'], loss: 0.0011874999999999993
['zatím'] into ['ten'], loss: 0.0011874999999999993
['skupiny'] into ['státní'], loss: 0.001187500000000001
['soudu'] into ['zahraničí'], loss: 0.0011884731162203135
['Praha'] into ['SSSR'], loss: 0.001243231658450781
['vůbec'] into ['mne'], loss: 0.0012478523202793827
['jenom'] into ['li'], loss: 0.0012767534522681188
['M'] into ['11'], loss: 0.0012816011987000966
['asi'] into ['zcela'], loss: 0.0012869489640353407
['Branky'] into ['kolo', 'OKD'], loss: 0.0013028219907537447
['především'] into ['zápas'], loss: 0.0013042915407808142
['svou'] into ['zase'], loss: 0.0013124999999999977
['ministr'] into ['vláda'], loss: 0.0013341294329205552
['Praze'] into ['době'], loss: 0.0013445499657345874
['kdy'] into ['kterou', 'protože'], loss: 0.001349505245968892
['Jan'] into ['prezident'], loss: 0.0013730391984779471
['Do'] into ['Podle'], loss: 0.0013863005917115045
['naší'] into ['sociální'], loss: 0.0013982538715143888
['5'] into ['7'], loss: 0.0014233355439576389
['přes'] into ['bez'], loss: 0.0014528195311147851
['listopadu'] into ['I', 'C'], loss: 0.0014884596631834987
['4'] into ['6'], loss: 0.001534306804530615
['mu'] into ['dosud'], loss: 0.0015560539092281831
['tomu'] into ['tím'], loss: 0.0015706982779089057
['které'] into ['kteří'], loss: 0.0015725560158424377
['státu'] into ['privatizace'], loss: 0.0016071509689703986
['všech'] into ['tohoto'], loss: 0.0016180571096380053
['dva'] into ['tento'], loss: 0.0016578940755229577
['ČR'] into ['&slash;'], loss: 0.001672987704790311
['měl'] into ['mohl', 'měla'], loss: 0.001715509125982959
['A'] into ['Je', 'Ale'], loss: 0.001716512552222207
['FS'] into ['nic'], loss: 0.001720770563345919
['již'] into ['NATO'], loss: 0.0017245186237007833
['jak'] into ['když'], loss: 0.0017280600113803654
['návrh'] into ['nové'], loss: 0.0017874551835489784
['tu'] into ['dnes'], loss: 0.0017905222142359347
['20'] into ['13'], loss: 0.0018034707099778904
['nad'] into ['tam'], loss: 0.001819391646208218
['tedy'] into ['ovšem'], loss: 0.0018506945430876503
['3'] into ['2'], loss: 0.0018552540832437459
['h'] into ['&percnt;', 'zájem'], loss: 0.0018791367416811974
['vlastní'] into ['tom'], loss: 0.0019014350043542252
['Jak', 'To'] into ['Když', 'Z'], loss: 0.001910405036593739
['ke'] into ['proti'], loss: 0.0019139097655573936
['strany'] into ['podniků', 'parlamentu'], loss: 0.0019218879806178236
['druhé', 'otázku'] into ['včerejší', 'sebe'], loss: 0.0019632050947216498
['ani'] into ['teď', 'všechno'], loss: 0.001972240404854938
['ho'] into ['pouze'], loss: 0.0020153195311147882
['1'] into ['0'], loss: 0.0020223970120349447
['KSČ', 'ES'] into ['ČNR', 'majetku'], loss: 0.002024736323118854
['mají'] into ['jim'], loss: 0.002037691124737481
['místo', 'pokud'] into ['později', 'třeba'], loss: 0.0020673210325877333
['1991', '18'] into ['10', '12'], loss: 0.0020825370164485463
['prý', 'přece'] into ['dokonce', 'často'], loss: 0.002083768634729254
['P', 'J'] into ['11', 'M'], loss: 0.0020955576043704665
['&ast;'] into ["'", 'Za'], loss: 0.0021106639025538837
['korun', 'dolarů'] into ['prostředí', 'lidí'], loss: 0.002127574621266941
['vlády'] into ['zahraničí', 'soudu'], loss: 0.002128546272704592
['jednání'] into ['roku'], loss: 0.0021529581516380753
['smlouvy', 'informace'] into ['poslanců', 'dnech'], loss: 0.0021628100413849036
['další'] into ['ceny', 'chce'], loss: 0.002179229296672174
['federální'] into ['její', 'tato'], loss: 0.00218634386765327
['PRAHA'] into ['státní', 'skupiny'], loss: 0.0022005546477683637
['který'] into ['kde', 'která'], loss: 0.002212449790603168
['let'] into ['Sparta', 'Moskva'], loss: 0.0022256015978940396
['roce', 'případě'] into ['době', 'Praze'], loss: 0.002246117768133918
['nebo'] into ['byla'], loss: 0.0022693720511780917
['st'] into ['nám', 'IFS'], loss: 0.002270056260488552
['války', 'organizace'] into ['SNR', 'den'], loss: 0.002272492847119026
['byly'] into ['ten', 'zatím'], loss: 0.002288473592426658
['české', 'nich'] into ['celé', 'České'], loss: 0.0022982933751332334
['co'] into ['než'], loss: 0.0023335544104494104
['programu', '25'] into ['mne', 'vůbec'], loss: 0.0023349905102701216
['není'] into ['li', 'jenom'], loss: 0.002341145571922235
['této', 'tuto'] into ['úterý', 'tomto'], loss: 0.002380316263194765
['První'] into ['Po', 'O'], loss: 0.0023890501922489843
['stavu', 'armády'] into ['zápas', 'především'], loss: 0.0024332046405587597
['těch', 'toho'] into ['lidé', 'proto'], loss: 0.002440635244398505
['poslední', 'jeden'] into ['některé', 'mít'], loss: 0.002475361200795871
['mi'] into ['zase', 'svou'], loss: 0.0024836785148541866
['má'] into ['musí'], loss: 0.0024836854819853416
['také'] into ['řekl'], loss: 0.0025000000000000014
['Kčs', 'K'] into ['I', 'C', 'listopadu'], loss: 0.0025591181329135786
['9', '8'] into ['7', '5'], loss: 0.0025791297045510177
['své'] into ['dvou', 'svých'], loss: 0.002582168631398502
['jsme'] into ['sám', 'může'], loss: 0.002686498045968969
['aby'] into ['kterou', 'protože', 'kdy'], loss: 0.0026945488277869524
['nás'] into ['zákona'], loss: 0.0027848688196145405
['vláda', 'ministr'] into ['prezident', 'Jan'], loss: 0.002812499999999999
['pak'] into ['tak'], loss: 0.0028285670774004373
['Na'] into ['Podle', 'Do'], loss: 0.002841520398666003
['&num;'] into ['či'], loss: 0.002874999999999999
['podle'] into ['zcela', 'asi'], loss: 0.0029923084529004766
['první'] into ['SSSR', 'Praha'], loss: 0.003081496340133173
['při'] into ['tisíc', 'miliónů'], loss: 0.003110557621313204
['ČSFR'] into ['&slash;', 'ČR'], loss: 0.003198068689439117
['žen'] into ['privatizace', 'státu'], loss: 0.0031999048303329204
['bez', 'přes'] into ['tohoto', 'všech'], loss: 0.003250000000000001
['sociální', 'naší'] into ['tento', 'dva'], loss: 0.0032606980528989728
['jejich'] into ['tom', 'vlastní'], loss: 0.0032759087023037647
['dosud', 'mu'] into ['jim', 'mají'], loss: 0.003325686046597299
['budou'] into ['nic', 'FS'], loss: 0.0034368063692512938
['mezi'] into ['proti', 'ke'], loss: 0.0034643783854885375
['='] into ["'", 'Za', '&ast;'], loss: 0.003572048714886117
['NATO', 'již'] into ['tam', 'nad'], loss: 0.003621115783858103
['ovšem', 'tedy'] into ['li', 'jenom', 'není'], loss: 0.0036925891408396676
['až'] into ['bylo'], loss: 0.0036975874695105142
['když', 'jak'] into ['později', 'třeba', 'místo', 'pokud'], loss: 0.0037618637569854004
['tím', 'tomu'] into ['řekl', 'také'], loss: 0.0037955665827846645
['teď', 'všechno', 'ani'] into ['dokonce', 'často', 'prý', 'přece'], loss: 0.0038807089931848122
['kolo', 'OKD', 'Branky'] into ['Když', 'Z', 'Jak', 'To'], loss: 0.0039197737909314415
['ještě'] into ['být'], loss: 0.003939529111445961
['utkání'] into ['nové', 'návrh'], loss: 0.003989063791411937
['10', '12', '1991', '18'] into ['7', '5', '9', '8'], loss: 0.004007448852547103
['však'] into ['dnes', 'tu'], loss: 0.004029196140738805
['kteří', 'které'] into ['než', 'co'], loss: 0.004064294100056022
['pouze', 'ho'] into ['zápas', 'především', 'stavu', 'armády'], loss: 0.004125448003320774
['podniků', 'parlamentu', 'strany'] into ['ČNR', 'majetku', 'KSČ', 'ES'], loss: 0.004211594545384369
['mohl', 'měla', 'měl'] into ['musí', 'má'], loss: 0.004272317958557526
['ceny', 'chce', 'další'] into ['SNR', 'den', 'války', 'organizace'], loss: 0.0043244147772177974
['Je', 'Ale', 'A'] into ['Po', 'O', 'První'], loss: 0.004380505874064937
['!'] into ['&percnt;', 'zájem', 'h'], loss: 0.004390974581898702
['13', '20'] into ['Sparta', 'Moskva', 'let'], loss: 0.004305709786259467
['roku', 'jednání'] into ['poslanců', 'dnech', 'smlouvy', 'informace'], loss: 0.004401178431686491
['už'] into ['ten', 'zatím', 'byly'], loss: 0.004485810154793851
['ze'] into [';'], loss: 0.004504381542021978
['prostředí', 'lidí', 'korun', 'dolarů'] into ['nám', 'IFS', 'st'], loss: 0.004507146340000337
['byl'] into ['byla', 'nebo'], loss: 0.004530639062229575
['včerejší', 'sebe', 'druhé', 'otázku'] into ['zákona', 'nás'], loss: 0.0045596289985178964
['její', 'tato', 'federální'] into ['některé', 'mít', 'poslední', 'jeden'], loss: 0.0045728614010567825
['mne', 'vůbec', 'programu', '25'] into ['lidé', 'proto', 'těch', 'toho'], loss: 0.004748937435083653
['včera'] into ['bude'], loss: 0.00477066148213461
['USA'] into ['SSSR', 'Praha', 'první'], loss: 0.004881459688664964
['u'] into ['jeho'], loss: 0.004905639062229577
['před'] into ['zase', 'svou', 'mi'], loss: 0.005147348011163736
['státní', 'skupiny', 'PRAHA'] into ['dvou', 'svých', 'své'], loss: 0.005156229027441242
['době', 'Praze', 'roce', 'případě'] into ['celé', 'České', 'české', 'nich'], loss: 0.005248800582776482
['jsem'] into ['sám', 'může', 'jsme'], loss: 0.00531076512416718
['jsou'] into ['zahraničí', 'soudu', 'vlády'], loss: 0.00538976639768567
['tak', 'pak'] into ['zcela', 'asi', 'podle'], loss: 0.005519516752857589
['ale'] into ['kterou', 'protože', 'kdy', 'aby'], loss: 0.005554489771549268
['?'] into ['prezident', 'Jan', 'vláda', 'ministr'], loss: 0.005600914324596355
['2', '3'] into ['7', '5', '9', '8', '10', '12', '1991', '18'], loss: 0.005832669813534676
['jen'] into ['či', '&num;'], loss: 0.0059238677664323084
['V'] into ['I', 'C', 'listopadu', 'Kčs', 'K'], loss: 0.005986487691836354
['že'] into ['kde', 'která', 'který'], loss: 0.0061651741129960524
['Podle', 'Do', 'Na'] into ['Když', 'Z', 'Jak', 'To', 'kolo', 'OKD', 'Branky'], loss: 0.006234922659799358
['od'] into ['tisíc', 'miliónů', 'při'], loss: 0.0062872295783028415
['úterý', 'tomto', 'této', 'tuto'] into ['celé', 'České', 'české', 'nich', 'době', 'Praze', 'roce', 'případě'], loss: 0.006349276116390923
['si'] into ['jim', 'mají', 'dosud', 'mu'], loss: 0.006408269572065707
['nic', 'FS', 'budou'] into ['dnes', 'tu', 'však'], loss: 0.006358295193085626
['6', '4'] into ['Sparta', 'Moskva', 'let', '13', '20'], loss: 0.00642369195787644
['tohoto', 'všech', 'bez', 'přes'] into ['tento', 'dva', 'sociální', 'naší'], loss: 0.0064544095628858685
['privatizace', 'státu', 'žen'] into ['tam', 'nad', 'NATO', 'již'], loss: 0.0067043694757753
['&slash;', 'ČR', 'ČSFR'] into ['ČNR', 'majetku', 'KSČ', 'ES', 'podniků', 'parlamentu', 'strany'], loss: 0.007022159823427105
['jako'] into ['proti', 'ke', 'mezi'], loss: 0.007099931528193331
['li', 'jenom', 'není', 'ovšem', 'tedy'] into ['bylo', 'až'], loss: 0.007194926685810704
['tom', 'vlastní', 'jejich'] into ['řekl', 'také', 'tím', 'tomu'], loss: 0.007353394626230476
['0', '1'] into ['7', '5', '9', '8', '10', '12', '1991', '18', '2', '3'], loss: 0.007692804369738099
['nové', 'návrh', 'utkání'] into [';', 'ze'], loss: 0.007426159012418433
['poslanců', 'dnech', 'smlouvy', 'informace', 'roku', 'jednání'] into ['zákona', 'nás', 'včerejší', 'sebe', 'druhé', 'otázku'], loss: 0.007787237212970951
['než', 'co', 'kteří', 'které'] into ['kterou', 'protože', 'kdy', 'aby', 'ale'], loss: 0.008036086658854122
['by'] into ['později', 'třeba', 'místo', 'pokud', 'když', 'jak'], loss: 0.008238247810362686
['za'] into ['být', 'ještě'], loss: 0.008434252877951954
["'", 'Za', '&ast;', '='] into ['&percnt;', 'zájem', 'h', '!'], loss: 0.00861886171779707
['SNR', 'den', 'války', 'organizace', 'ceny', 'chce', 'další'] into ['zahraničí', 'soudu', 'vlády', 'jsou'], loss: 0.008760138873449282
['pro'] into ['zápas', 'především', 'stavu', 'armády', 'pouze', 'ho'], loss: 0.008975021669750794
['byla', 'nebo', 'byl'] into ['některé', 'mít', 'poslední', 'jeden', 'její', 'tato', 'federální'], loss: 0.00896437368103599
['ten', 'zatím', 'byly', 'už'] into ['bude', 'včera'], loss: 0.008954397905722129
['dokonce', 'často', 'prý', 'přece', 'teď', 'všechno', 'ani'] into ['zase', 'svou', 'mi', 'před'], loss: 0.008970100646232286
['ve'] into ['po'], loss: 0.009005051977045125
['Po', 'O', 'První', 'Je', 'Ale', 'A'] into ['Když', 'Z', 'Jak', 'To', 'kolo', 'OKD', 'Branky', 'Podle', 'Do', 'Na'], loss: 0.009298121255074787
['11', 'M', 'P', 'J'] into ['Sparta', 'Moskva', 'let', '13', '20', '6', '4'], loss: 0.009681760451150025
['musí', 'má', 'mohl', 'měla', 'měl'] into ['sám', 'může', 'jsme', 'jsem'], loss: 0.009818447253199405
['lidé', 'proto', 'těch', 'toho', 'mne', 'vůbec', 'programu', '25'] into ['jeho', 'u'], loss: 0.009992553249331947
['dvou', 'svých', 'své', 'státní', 'skupiny', 'PRAHA'] into ['prezident', 'Jan', 'vláda', 'ministr', '?'], loss: 0.010497776232287294
['nám', 'IFS', 'st', 'prostředí', 'lidí', 'korun', 'dolarů'] into ['tam', 'nad', 'NATO', 'již', 'privatizace', 'státu', 'žen'], loss: 0.01066070296083086
['zcela', 'asi', 'podle', 'tak', 'pak'] into ['dnes', 'tu', 'však', 'nic', 'FS', 'budou'], loss: 0.01083856083390548
['SSSR', 'Praha', 'první', 'USA'] into ['ČNR', 'majetku', 'KSČ', 'ES', 'podniků', 'parlamentu', 'strany', '&slash;', 'ČR', 'ČSFR'], loss: 0.011179118962051914
['k'] into ['tisíc', 'miliónů', 'při', 'od'], loss: 0.011270399834148511
['to'] into ['řekl', 'také', 'tím', 'tomu', 'tom', 'vlastní', 'jejich'], loss: 0.011767899265340743
['či', '&num;', 'jen'] into ['bylo', 'až', 'li', 'jenom', 'není', 'ovšem', 'tedy'], loss: 0.012159785666876009
['i'] into ['jim', 'mají', 'dosud', 'mu', 'si'], loss: 0.012430601628855549
['z'] into ['tento', 'dva', 'sociální', 'naší', 'tohoto', 'všech', 'bez', 'přes'], loss: 0.013879357773197236
['I', 'C', 'listopadu', 'Kčs', 'K', 'V'] into ['Sparta', 'Moskva', 'let', '13', '20', '6', '4', '11', 'M', 'P', 'J'], loss: 0.01391457361069619
['je'] into ['později', 'třeba', 'místo', 'pokud', 'když', 'jak', 'by'], loss: 0.014102668905790323
['proti', 'ke', 'mezi', 'jako'] into ['být', 'ještě', 'za'], loss: 0.0142646325516133
['bude', 'včera', 'ten', 'zatím', 'byly', 'už'] into ['zase', 'svou', 'mi', 'před', 'dokonce', 'často', 'prý', 'přece', 'teď', 'všechno', 'ani'], loss: 0.015050366376280376
['celé', 'České', 'české', 'nich', 'době', 'Praze', 'roce', 'případě', 'úterý', 'tomto', 'této', 'tuto'] into ['zahraničí', 'soudu', 'vlády', 'jsou', 'SNR', 'den', 'války', 'organizace', 'ceny', 'chce', 'další'], loss: 0.015959838978526125
['do'] into [';', 'ze', 'nové', 'návrh', 'utkání'], loss: 0.01615810372473548
['některé', 'mít', 'poslední', 'jeden', 'její', 'tato', 'federální', 'byla', 'nebo', 'byl'] into ['jeho', 'u', 'lidé', 'proto', 'těch', 'toho', 'mne', 'vůbec', 'programu', '25'], loss: 0.016458803243523872
['zápas', 'především', 'stavu', 'armády', 'pouze', 'ho', 'pro'] into ['sám', 'může', 'jsme', 'jsem', 'musí', 'má', 'mohl', 'měla', 'měl'], loss: 0.016829705071551262
['s'] into ['po', 've'], loss: 0.017263915207179747
['zákona', 'nás', 'včerejší', 'sebe', 'druhé', 'otázku', 'poslanců', 'dnech', 'smlouvy', 'informace', 'roku', 'jednání'] into ['&percnt;', 'zájem', 'h', '!', "'", 'Za', '&ast;', '='], loss: 0.018178601896546787
[')'] into ['tam', 'nad', 'NATO', 'již', 'privatizace', 'státu', 'žen', 'nám', 'IFS', 'st', 'prostředí', 'lidí', 'korun', 'dolarů'], loss: 0.019597243864837143
['dnes', 'tu', 'však', 'nic', 'FS', 'budou', 'zcela', 'asi', 'podle', 'tak', 'pak'] into ['řekl', 'také', 'tím', 'tomu', 'tom', 'vlastní', 'jejich', 'to'], loss: 0.02019180581431627
['prezident', 'Jan', 'vláda', 'ministr', '?', 'dvou', 'svých', 'své', 'státní', 'skupiny', 'PRAHA'] into ['tisíc', 'miliónů', 'při', 'od', 'k'], loss: 0.02132934213964802
['ČNR', 'majetku', 'KSČ', 'ES', 'podniků', 'parlamentu', 'strany', '&slash;', 'ČR', 'ČSFR', 'SSSR', 'Praha', 'první', 'USA'] into ['bylo', 'až', 'li', 'jenom', 'není', 'ovšem', 'tedy', 'či', '&num;', 'jen'], loss: 0.022629925937956243
['kterou', 'protože', 'kdy', 'aby', 'ale', 'než', 'co', 'kteří', 'které'] into ['později', 'třeba', 'místo', 'pokud', 'když', 'jak', 'by', 'je'], loss: 0.022996237778382708
['kde', 'která', 'který', 'že'] into ['později', 'třeba', 'místo', 'pokud', 'když', 'jak', 'by', 'je', 'kterou', 'protože', 'kdy', 'aby', 'ale', 'než', 'co', 'kteří', 'které'], loss: 0.02133642361451281
['jim', 'mají', 'dosud', 'mu', 'si', 'i'] into ['být', 'ještě', 'za', 'proti', 'ke', 'mezi', 'jako'], loss: 0.024187697600952917
['o'] into ['tento', 'dva', 'sociální', 'naší', 'tohoto', 'všech', 'bez', 'přes', 'z'], loss: 0.024565154887369715
['7', '5', '9', '8', '10', '12', '1991', '18', '2', '3', '0', '1'] into ['Sparta', 'Moskva', 'let', '13', '20', '6', '4', '11', 'M', 'P', 'J', 'I', 'C', 'listopadu', 'Kčs', 'K', 'V'], loss: 0.02517990273816964
[':'] into ['('], loss: 0.026286911014724034
['"'] into ['Když', 'Z', 'Jak', 'To', 'kolo', 'OKD', 'Branky', 'Podle', 'Do', 'Na', 'Po', 'O', 'První', 'Je', 'Ale', 'A'], loss: 0.026936736442399353
['zahraničí', 'soudu', 'vlády', 'jsou', 'SNR', 'den', 'války', 'organizace', 'ceny', 'chce', 'další', 'celé', 'České', 'české', 'nich', 'době', 'Praze', 'roce', 'případě', 'úterý', 'tomto', 'této', 'tuto'] into ['&percnt;', 'zájem', 'h', '!', "'", 'Za', '&ast;', '=', 'zákona', 'nás', 'včerejší', 'sebe', 'druhé', 'otázku', 'poslanců', 'dnech', 'smlouvy', 'informace', 'roku', 'jednání'], loss: 0.02742165251266529
['zase', 'svou', 'mi', 'před', 'dokonce', 'často', 'prý', 'přece', 'teď', 'všechno', 'ani', 'bude', 'včera', 'ten', 'zatím', 'byly', 'už'] into ['sám', 'může', 'jsme', 'jsem', 'musí', 'má', 'mohl', 'měla', 'měl', 'zápas', 'především', 'stavu', 'armády', 'pouze', 'ho', 'pro'], loss: 0.027985776249456966
['jeho', 'u', 'lidé', 'proto', 'těch', 'toho', 'mne', 'vůbec', 'programu', '25', 'některé', 'mít', 'poslední', 'jeden', 'její', 'tato', 'federální', 'byla', 'nebo', 'byl'] into ['řekl', 'také', 'tím', 'tomu', 'tom', 'vlastní', 'jejich', 'to', 'dnes', 'tu', 'však', 'nic', 'FS', 'budou', 'zcela', 'asi', 'podle', 'tak', 'pak'], loss: 0.02920638102938234
[';', 'ze', 'nové', 'návrh', 'utkání', 'do'] into ['po', 've', 's'], loss: 0.03126716129684685
['na'] into ['tisíc', 'miliónů', 'při', 'od', 'k', 'prezident', 'Jan', 'vláda', 'ministr', '?', 'dvou', 'svých', 'své', 'státní', 'skupiny', 'PRAHA'], loss: 0.03698270723703551
['tam', 'nad', 'NATO', 'již', 'privatizace', 'státu', 'žen', 'nám', 'IFS', 'st', 'prostředí', 'lidí', 'korun', 'dolarů', ')'] into ['bylo', 'až', 'li', 'jenom', 'není', 'ovšem', 'tedy', 'či', '&num;', 'jen', 'ČNR', 'majetku', 'KSČ', 'ES', 'podniků', 'parlamentu', 'strany', '&slash;', 'ČR', 'ČSFR', 'SSSR', 'Praha', 'první', 'USA'], loss: 0.03914143532004303
['se'] into ['být', 'ještě', 'za', 'proti', 'ke', 'mezi', 'jako', 'jim', 'mají', 'dosud', 'mu', 'si', 'i'], loss: 0.042315502042312256
['v'] into ['tento', 'dva', 'sociální', 'naší', 'tohoto', 'všech', 'bez', 'přes', 'z', 'o'], loss: 0.04965256852182998
['-'] into ['(', ':'], loss: 0.050608285188534705
1: ['C'] into ['I']
2: ['Ale'] into ['Je']
3: ['To'] into ['Jak']
4: ['J'] into ['P']
5: ['OKD'] into ['kolo']
6: ['protože'] into ['kterou']
7: ['pokud'] into ['místo']
8: ['ES'] into ['KSČ']
9: ['otázku'] into ['druhé']
10: ['Moskva'] into ['Sparta']
11: ['parlamentu'] into ['podniků']
12: ['toho'] into ['těch']
13: ['18'] into ['1991']
14: ['třeba'] into ['později']
15: ['Z'] into ['Když']
16: ['měla'] into ['mohl']
17: ['dolarů'] into ['korun']
18: ['přece'] into ['prý']
19: ['informace'] into ['smlouvy']
20: ['8'] into ['9']
21: ['sebe'] into ['včerejší']
22: ['často'] into ['dokonce']
23: ['25'] into ['programu']
24: ['organizace'] into ['války']
25: ['lidí'] into ['prostředí']
26: ['případě'] into ['roce']
27: ['všechno'] into ['teď']
28: ['jeden'] into ['poslední']
29: ['nich'] into ['české']
30: ['IFS'] into ['nám']
31: ['den'] into ['SNR']
32: ['tuto'] into ['této']
33: ['tato'] into ['její']
34: ['armády'] into ['stavu']
35: ['Za'] into ["'"]
36: ['svých'] into ['dvou']
37: ['České'] into ['celé']
38: ['zájem'] into ['&percnt;']
39: ['12'] into ['10']
40: ['proto'] into ['lidé']
41: ['dnech'] into ['poslanců']
42: ['miliónů'] into ['tisíc']
43: ['K'] into ['Kčs']
44: ['tomto'] into ['úterý']
45: ['majetku'] into ['ČNR']
46: ['O'] into ['Po']
47: ['která'] into ['kde']
48: ['mít'] into ['některé']
49: ['chce'] into ['ceny']
50: ['může'] into ['sám']
51: ['zatím'] into ['ten']
52: ['skupiny'] into ['státní']
53: ['soudu'] into ['zahraničí']
54: ['Praha'] into ['SSSR']
55: ['vůbec'] into ['mne']
56: ['jenom'] into ['li']
57: ['M'] into ['11']
58: ['asi'] into ['zcela']
59: ['Branky'] into ['kolo', 'OKD']
60: ['především'] into ['zápas']
61: ['svou'] into ['zase']
62: ['ministr'] into ['vláda']
63: ['Praze'] into ['době']
64: ['kdy'] into ['kterou', 'protože']
65: ['Jan'] into ['prezident']
66: ['Do'] into ['Podle']
67: ['naší'] into ['sociální']
68: ['5'] into ['7']
69: ['přes'] into ['bez']
70: ['listopadu'] into ['I', 'C']
71: ['4'] into ['6']
72: ['mu'] into ['dosud']
73: ['tomu'] into ['tím']
74: ['které'] into ['kteří']
75: ['státu'] into ['privatizace']
76: ['všech'] into ['tohoto']
77: ['dva'] into ['tento']
78: ['ČR'] into ['&slash;']
79: ['měl'] into ['mohl', 'měla']
80: ['A'] into ['Je', 'Ale']
81: ['FS'] into ['nic']
82: ['již'] into ['NATO']
83: ['jak'] into ['když']
84: ['návrh'] into ['nové']
85: ['tu'] into ['dnes']
86: ['20'] into ['13']
87: ['nad'] into ['tam']
88: ['tedy'] into ['ovšem']
89: ['3'] into ['2']
90: ['h'] into ['&percnt;', 'zájem']
91: ['vlastní'] into ['tom']
92: ['Jak', 'To'] into ['Když', 'Z']
93: ['ke'] into ['proti']
94: ['strany'] into ['podniků', 'parlamentu']
95: ['druhé', 'otázku'] into ['včerejší', 'sebe']
96: ['ani'] into ['teď', 'všechno']
97: ['ho'] into ['pouze']
98: ['1'] into ['0']
99: ['KSČ', 'ES'] into ['ČNR', 'majetku']
100: ['mají'] into ['jim']
101: ['místo', 'pokud'] into ['později', 'třeba']
102: ['1991', '18'] into ['10', '12']
103: ['prý', 'přece'] into ['dokonce', 'často']
104: ['P', 'J'] into ['11', 'M']
105: ['&ast;'] into ["'", 'Za']
106: ['korun', 'dolarů'] into ['prostředí', 'lidí']
107: ['vlády'] into ['zahraničí', 'soudu']
108: ['jednání'] into ['roku']
109: ['smlouvy', 'informace'] into ['poslanců', 'dnech']
110: ['další'] into ['ceny', 'chce']
111: ['federální'] into ['její', 'tato']
112: ['PRAHA'] into ['státní', 'skupiny']
113: ['který'] into ['kde', 'která']
114: ['let'] into ['Sparta', 'Moskva']
115: ['roce', 'případě'] into ['době', 'Praze']
116: ['nebo'] into ['byla']
117: ['st'] into ['nám', 'IFS']
118: ['války', 'organizace'] into ['SNR', 'den']
119: ['byly'] into ['ten', 'zatím']
120: ['české', 'nich'] into ['celé', 'České']
121: ['co'] into ['než']
122: ['programu', '25'] into ['mne', 'vůbec']
123: ['není'] into ['li', 'jenom']
124: ['této', 'tuto'] into ['úterý', 'tomto']
125: ['První'] into ['Po', 'O']
126: ['stavu', 'armády'] into ['zápas', 'především']
127: ['těch', 'toho'] into ['lidé', 'proto']
128: ['poslední', 'jeden'] into ['některé', 'mít']
129: ['mi'] into ['zase', 'svou']
130: ['má'] into ['musí']
131: ['také'] into ['řekl']
132: ['Kčs', 'K'] into ['I', 'C', 'listopadu']
133: ['9', '8'] into ['7', '5']
134: ['své'] into ['dvou', 'svých']
135: ['jsme'] into ['sám', 'může']
136: ['aby'] into ['kterou', 'protože', 'kdy']
137: ['nás'] into ['zákona']
138: ['vláda', 'ministr'] into ['prezident', 'Jan']
139: ['pak'] into ['tak']
140: ['Na'] into ['Podle', 'Do']
141: ['&num;'] into ['či']
142: ['podle'] into ['zcela', 'asi']
143: ['první'] into ['SSSR', 'Praha']
144: ['při'] into ['tisíc', 'miliónů']
145: ['ČSFR'] into ['&slash;', 'ČR']
146: ['žen'] into ['privatizace', 'státu']
147: ['bez', 'přes'] into ['tohoto', 'všech']
148: ['sociální', 'naší'] into ['tento', 'dva']
149: ['jejich'] into ['tom', 'vlastní']
150: ['dosud', 'mu'] into ['jim', 'mají']
151: ['budou'] into ['nic', 'FS']
152: ['mezi'] into ['proti', 'ke']
153: ['='] into ["'", 'Za', '&ast;']
154: ['NATO', 'již'] into ['tam', 'nad']
155: ['ovšem', 'tedy'] into ['li', 'jenom', 'není']
156: ['až'] into ['bylo']
157: ['když', 'jak'] into ['později', 'třeba', 'místo', 'pokud']
158: ['tím', 'tomu'] into ['řekl', 'také']
159: ['teď', 'všechno', 'ani'] into ['dokonce', 'často', 'prý', 'přece']
160: ['kolo', 'OKD', 'Branky'] into ['Když', 'Z', 'Jak', 'To']
161: ['ještě'] into ['být']
162: ['utkání'] into ['nové', 'návrh']
163: ['10', '12', '1991', '18'] into ['7', '5', '9', '8']
164: ['však'] into ['dnes', 'tu']
165: ['kteří', 'které'] into ['než', 'co']
166: ['pouze', 'ho'] into ['zápas', 'především', 'stavu', 'armády']
167: ['podniků', 'parlamentu', 'strany'] into ['ČNR', 'majetku', 'KSČ', 'ES']
168: ['mohl', 'měla', 'měl'] into ['musí', 'má']
169: ['ceny', 'chce', 'další'] into ['SNR', 'den', 'války', 'organizace']
170: ['Je', 'Ale', 'A'] into ['Po', 'O', 'První']
171: ['!'] into ['&percnt;', 'zájem', 'h']
172: ['13', '20'] into ['Sparta', 'Moskva', 'let']
173: ['roku', 'jednání'] into ['poslanců', 'dnech', 'smlouvy', 'informace']
174: ['už'] into ['ten', 'zatím', 'byly']
175: ['ze'] into [';']
176: ['prostředí', 'lidí', 'korun', 'dolarů'] into ['nám', 'IFS', 'st']
177: ['byl'] into ['byla', 'nebo']
178: ['včerejší', 'sebe', 'druhé', 'otázku'] into ['zákona', 'nás']
179: ['její', 'tato', 'federální'] into ['některé', 'mít', 'poslední', 'jeden']
180: ['mne', 'vůbec', 'programu', '25'] into ['lidé', 'proto', 'těch', 'toho']
181: ['včera'] into ['bude']
182: ['USA'] into ['SSSR', 'Praha', 'první']
183: ['u'] into ['jeho']
184: ['před'] into ['zase', 'svou', 'mi']
185: ['státní', 'skupiny', 'PRAHA'] into ['dvou', 'svých', 'své']
186: ['době', 'Praze', 'roce', 'případě'] into ['celé', 'České', 'české', 'nich']
187: ['jsem'] into ['sám', 'může', 'jsme']
188: ['jsou'] into ['zahraničí', 'soudu', 'vlády']
189: ['tak', 'pak'] into ['zcela', 'asi', 'podle']
190: ['ale'] into ['kterou', 'protože', 'kdy', 'aby']
191: ['?'] into ['prezident', 'Jan', 'vláda', 'ministr']
192: ['2', '3'] into ['7', '5', '9', '8', '10', '12', '1991', '18']
193: ['jen'] into ['či', '&num;']
194: ['V'] into ['I', 'C', 'listopadu', 'Kčs', 'K']
195: ['že'] into ['kde', 'která', 'který']
196: ['Podle', 'Do', 'Na'] into ['Když', 'Z', 'Jak', 'To', 'kolo', 'OKD', 'Branky']
197: ['od'] into ['tisíc', 'miliónů', 'při']
198: ['úterý', 'tomto', 'této', 'tuto'] into ['celé', 'České', 'české', 'nich', 'době', 'Praze', 'roce', 'případě']
199: ['si'] into ['jim', 'mají', 'dosud', 'mu']
200: ['nic', 'FS', 'budou'] into ['dnes', 'tu', 'však']
201: ['6', '4'] into ['Sparta', 'Moskva', 'let', '13', '20']
202: ['tohoto', 'všech', 'bez', 'přes'] into ['tento', 'dva', 'sociální', 'naší']
203: ['privatizace', 'státu', 'žen'] into ['tam', 'nad', 'NATO', 'již']
204: ['&slash;', 'ČR', 'ČSFR'] into ['ČNR', 'majetku', 'KSČ', 'ES', 'podniků', 'parlamentu', 'strany']
205: ['jako'] into ['proti', 'ke', 'mezi']
206: ['li', 'jenom', 'není', 'ovšem', 'tedy'] into ['bylo', 'až']
207: ['tom', 'vlastní', 'jejich'] into ['řekl', 'také', 'tím', 'tomu']
208: ['0', '1'] into ['7', '5', '9', '8', '10', '12', '1991', '18', '2', '3']
209: ['nové', 'návrh', 'utkání'] into [';', 'ze']
210: ['poslanců', 'dnech', 'smlouvy', 'informace', 'roku', 'jednání'] into ['zákona', 'nás', 'včerejší', 'sebe', 'druhé', 'otázku']
211: ['než', 'co', 'kteří', 'které'] into ['kterou', 'protože', 'kdy', 'aby', 'ale']
212: ['by'] into ['později', 'třeba', 'místo', 'pokud', 'když', 'jak']
213: ['za'] into ['být', 'ještě']
214: ["'", 'Za', '&ast;', '='] into ['&percnt;', 'zájem', 'h', '!']
215: ['SNR', 'den', 'války', 'organizace', 'ceny', 'chce', 'další'] into ['zahraničí', 'soudu', 'vlády', 'jsou']
216: ['pro'] into ['zápas', 'především', 'stavu', 'armády', 'pouze', 'ho']
217: ['byla', 'nebo', 'byl'] into ['některé', 'mít', 'poslední', 'jeden', 'její', 'tato', 'federální']
218: ['ten', 'zatím', 'byly', 'už'] into ['bude', 'včera']
219: ['dokonce', 'často', 'prý', 'přece', 'teď', 'všechno', 'ani'] into ['zase', 'svou', 'mi', 'před']
220: ['ve'] into ['po']
221: ['Po', 'O', 'První', 'Je', 'Ale', 'A'] into ['Když', 'Z', 'Jak', 'To', 'kolo', 'OKD', 'Branky', 'Podle', 'Do', 'Na']
222: ['11', 'M', 'P', 'J'] into ['Sparta', 'Moskva', 'let', '13', '20', '6', '4']
223: ['musí', 'má', 'mohl', 'měla', 'měl'] into ['sám', 'může', 'jsme', 'jsem']
224: ['lidé', 'proto', 'těch', 'toho', 'mne', 'vůbec', 'programu', '25'] into ['jeho', 'u']
225: ['dvou', 'svých', 'své', 'státní', 'skupiny', 'PRAHA'] into ['prezident', 'Jan', 'vláda', 'ministr', '?']
226: ['nám', 'IFS', 'st', 'prostředí', 'lidí', 'korun', 'dolarů'] into ['tam', 'nad', 'NATO', 'již', 'privatizace', 'státu', 'žen']
227: ['zcela', 'asi', 'podle', 'tak', 'pak'] into ['dnes', 'tu', 'však', 'nic', 'FS', 'budou']
228: ['SSSR', 'Praha', 'první', 'USA'] into ['ČNR', 'majetku', 'KSČ', 'ES', 'podniků', 'parlamentu', 'strany', '&slash;', 'ČR', 'ČSFR']
229: ['k'] into ['tisíc', 'miliónů', 'při', 'od']
230: ['to'] into ['řekl', 'také', 'tím', 'tomu', 'tom', 'vlastní', 'jejich']
231: ['či', '&num;', 'jen'] into ['bylo', 'až', 'li', 'jenom', 'není', 'ovšem', 'tedy']
232: ['i'] into ['jim', 'mají', 'dosud', 'mu', 'si']
233: ['z'] into ['tento', 'dva', 'sociální', 'naší', 'tohoto', 'všech', 'bez', 'přes']
234: ['I', 'C', 'listopadu', 'Kčs', 'K', 'V'] into ['Sparta', 'Moskva', 'let', '13', '20', '6', '4', '11', 'M', 'P', 'J']
235: ['je'] into ['později', 'třeba', 'místo', 'pokud', 'když', 'jak', 'by']
236: ['proti', 'ke', 'mezi', 'jako'] into ['být', 'ještě', 'za']
237: ['bude', 'včera', 'ten', 'zatím', 'byly', 'už'] into ['zase', 'svou', 'mi', 'před', 'dokonce', 'často', 'prý', 'přece', 'teď', 'všechno', 'ani']
238: ['celé', 'České', 'české', 'nich', 'době', 'Praze', 'roce', 'případě', 'úterý', 'tomto', 'této', 'tuto'] into ['zahraničí', 'soudu', 'vlády', 'jsou', 'SNR', 'den', 'války', 'organizace', 'ceny', 'chce', 'další']
239: ['do'] into [';', 'ze', 'nové', 'návrh', 'utkání']
240: ['některé', 'mít', 'poslední', 'jeden', 'její', 'tato', 'federální', 'byla', 'nebo', 'byl'] into ['jeho', 'u', 'lidé', 'proto', 'těch', 'toho', 'mne', 'vůbec', 'programu', '25']
241: ['zápas', 'především', 'stavu', 'armády', 'pouze', 'ho', 'pro'] into ['sám', 'může', 'jsme', 'jsem', 'musí', 'má', 'mohl', 'měla', 'měl']
242: ['s'] into ['po', 've']
243: ['zákona', 'nás', 'včerejší', 'sebe', 'druhé', 'otázku', 'poslanců', 'dnech', 'smlouvy', 'informace', 'roku', 'jednání'] into ['&percnt;', 'zájem', 'h', '!', "'", 'Za', '&ast;', '=']
244: [')'] into ['tam', 'nad', 'NATO', 'již', 'privatizace', 'státu', 'žen', 'nám', 'IFS', 'st', 'prostředí', 'lidí', 'korun', 'dolarů']
245: ['dnes', 'tu', 'však', 'nic', 'FS', 'budou', 'zcela', 'asi', 'podle', 'tak', 'pak'] into ['řekl', 'také', 'tím', 'tomu', 'tom', 'vlastní', 'jejich', 'to']
246: ['prezident', 'Jan', 'vláda', 'ministr', '?', 'dvou', 'svých', 'své', 'státní', 'skupiny', 'PRAHA'] into ['tisíc', 'miliónů', 'při', 'od', 'k']
247: ['ČNR', 'majetku', 'KSČ', 'ES', 'podniků', 'parlamentu', 'strany', '&slash;', 'ČR', 'ČSFR', 'SSSR', 'Praha', 'první', 'USA'] into ['bylo', 'až', 'li', 'jenom', 'není', 'ovšem', 'tedy', 'či', '&num;', 'jen']
248: ['kterou', 'protože', 'kdy', 'aby', 'ale', 'než', 'co', 'kteří', 'které'] into ['později', 'třeba', 'místo', 'pokud', 'když', 'jak', 'by', 'je']
249: ['kde', 'která', 'který', 'že'] into ['později', 'třeba', 'místo', 'pokud', 'když', 'jak', 'by', 'je', 'kterou', 'protože', 'kdy', 'aby', 'ale', 'než', 'co', 'kteří', 'které']
250: ['jim', 'mají', 'dosud', 'mu', 'si', 'i'] into ['být', 'ještě', 'za', 'proti', 'ke', 'mezi', 'jako']
251: ['o'] into ['tento', 'dva', 'sociální', 'naší', 'tohoto', 'všech', 'bez', 'přes', 'z']
252: ['7', '5', '9', '8', '10', '12', '1991', '18', '2', '3', '0', '1'] into ['Sparta', 'Moskva', 'let', '13', '20', '6', '4', '11', 'M', 'P', 'J', 'I', 'C', 'listopadu', 'Kčs', 'K', 'V']
253: [':'] into ['(']
254: ['"'] into ['Když', 'Z', 'Jak', 'To', 'kolo', 'OKD', 'Branky', 'Podle', 'Do', 'Na', 'Po', 'O', 'První', 'Je', 'Ale', 'A']
255: ['zahraničí', 'soudu', 'vlády', 'jsou', 'SNR', 'den', 'války', 'organizace', 'ceny', 'chce', 'další', 'celé', 'České', 'české', 'nich', 'době', 'Praze', 'roce', 'případě', 'úterý', 'tomto', 'této', 'tuto'] into ['&percnt;', 'zájem', 'h', '!', "'", 'Za', '&ast;', '=', 'zákona', 'nás', 'včerejší', 'sebe', 'druhé', 'otázku', 'poslanců', 'dnech', 'smlouvy', 'informace', 'roku', 'jednání']
256: ['zase', 'svou', 'mi', 'před', 'dokonce', 'často', 'prý', 'přece', 'teď', 'všechno', 'ani', 'bude', 'včera', 'ten', 'zatím', 'byly', 'už'] into ['sám', 'může', 'jsme', 'jsem', 'musí', 'má', 'mohl', 'měla', 'měl', 'zápas', 'především', 'stavu', 'armády', 'pouze', 'ho', 'pro']
257: ['jeho', 'u', 'lidé', 'proto', 'těch', 'toho', 'mne', 'vůbec', 'programu', '25', 'některé', 'mít', 'poslední', 'jeden', 'její', 'tato', 'federální', 'byla', 'nebo', 'byl'] into ['řekl', 'také', 'tím', 'tomu', 'tom', 'vlastní', 'jejich', 'to', 'dnes', 'tu', 'však', 'nic', 'FS', 'budou', 'zcela', 'asi', 'podle', 'tak', 'pak']
258: [';', 'ze', 'nové', 'návrh', 'utkání', 'do'] into ['po', 've', 's']
259: ['na'] into ['tisíc', 'miliónů', 'při', 'od', 'k', 'prezident', 'Jan', 'vláda', 'ministr', '?', 'dvou', 'svých', 'své', 'státní', 'skupiny', 'PRAHA']
260: ['tam', 'nad', 'NATO', 'již', 'privatizace', 'státu', 'žen', 'nám', 'IFS', 'st', 'prostředí', 'lidí', 'korun', 'dolarů', ')'] into ['bylo', 'až', 'li', 'jenom', 'není', 'ovšem', 'tedy', 'či', '&num;', 'jen', 'ČNR', 'majetku', 'KSČ', 'ES', 'podniků', 'parlamentu', 'strany', '&slash;', 'ČR', 'ČSFR', 'SSSR', 'Praha', 'první', 'USA']
261: ['se'] into ['být', 'ještě', 'za', 'proti', 'ke', 'mezi', 'jako', 'jim', 'mají', 'dosud', 'mu', 'si', 'i']
262: ['v'] into ['tento', 'dva', 'sociální', 'naší', 'tohoto', 'všech', 'bez', 'přes', 'z', 'o']
263: ['-'] into ['(', ':']
Class 1: ['bylo', 'až', 'li', 'jenom', 'není', 'ovšem', 'tedy', 'či', '&num;', 'jen', 'ČNR', 'majetku', 'KSČ', 'ES', 'podniků', 'parlamentu', 'strany', '&slash;', 'ČR', 'ČSFR', 'SSSR', 'Praha', 'první', 'USA', 'tam', 'nad', 'NATO', 'již', 'privatizace', 'státu', 'žen', 'nám', 'IFS', 'st', 'prostředí', 'lidí', 'korun', 'dolarů', ')']
Class 2: ['později', 'třeba', 'místo', 'pokud', 'když', 'jak', 'by', 'je', 'kterou', 'protože', 'kdy', 'aby', 'ale', 'než', 'co', 'kteří', 'které', 'kde', 'která', 'který', 'že']
Class 3: ['Když', 'Z', 'Jak', 'To', 'kolo', 'OKD', 'Branky', 'Podle', 'Do', 'Na', 'Po', 'O', 'První', 'Je', 'Ale', 'A', '"']
Class 4: ['Sparta', 'Moskva', 'let', '13', '20', '6', '4', '11', 'M', 'P', 'J', 'I', 'C', 'listopadu', 'Kčs', 'K', 'V', '7', '5', '9', '8', '10', '12', '1991', '18', '2', '3', '0', '1']
Class 5: ['&percnt;', 'zájem', 'h', '!', "'", 'Za', '&ast;', '=', 'zákona', 'nás', 'včerejší', 'sebe', 'druhé', 'otázku', 'poslanců', 'dnech', 'smlouvy', 'informace', 'roku', 'jednání', 'zahraničí', 'soudu', 'vlády', 'jsou', 'SNR', 'den', 'války', 'organizace', 'ceny', 'chce', 'další', 'celé', 'České', 'české', 'nich', 'době', 'Praze', 'roce', 'případě', 'úterý', 'tomto', 'této', 'tuto']
Class 6: ['tisíc', 'miliónů', 'při', 'od', 'k', 'prezident', 'Jan', 'vláda', 'ministr', '?', 'dvou', 'svých', 'své', 'státní', 'skupiny', 'PRAHA', 'na']
Class 7: ['tento', 'dva', 'sociální', 'naší', 'tohoto', 'všech', 'bez', 'přes', 'z', 'o', 'v']
Class 8: ['sám', 'může', 'jsme', 'jsem', 'musí', 'má', 'mohl', 'měla', 'měl', 'zápas', 'především', 'stavu', 'armády', 'pouze', 'ho', 'pro', 'zase', 'svou', 'mi', 'před', 'dokonce', 'často', 'prý', 'přece', 'teď', 'všechno', 'ani', 'bude', 'včera', 'ten', 'zatím', 'byly', 'už']
Class 9: ['po', 've', 's', ';', 'ze', 'nové', 'návrh', 'utkání', 'do']
Class 10: ['řekl', 'také', 'tím', 'tomu', 'tom', 'vlastní', 'jejich', 'to', 'dnes', 'tu', 'však', 'nic', 'FS', 'budou', 'zcela', 'asi', 'podle', 'tak', 'pak', 'jeho', 'u', 'lidé', 'proto', 'těch', 'toho', 'mne', 'vůbec', 'programu', '25', 'některé', 'mít', 'poslední', 'jeden', 'její', 'tato', 'federální', 'byla', 'nebo', 'byl']
Class 11: ['(', ':', '-']
Class 12: ['být', 'ještě', 'za', 'proti', 'ke', 'mezi', 'jako', 'jim', 'mají', 'dosud', 'mu', 'si', 'i', 'se']
Class 13: ['a']
Class 14: ['.']
Class 15: [',']


Even though it is very hard to measure whether the merging is performing better then before, it optically seems little more promising, especially at the beginning (first 20 merges).

Merging tags¶

Since there are less tags than words, the algorithm works far quicker and allows to run on longer texts. Some interesting final classes are:

  • Class 1: ['VBZ', 'VBP', 'VB'] - representing present tense verbs
  • Class 8: [',', ':', '.', '('] - representing all functional characters
  • Class 2: ['TO', 'MD'] - representing modal verbs and "to" ('to go'...)

Final remark¶

Technical notes¶

All the histories of merges can be found in the result directory. The naming convention is the following:

{language}_{type}_{final_number_of_classes}_history.txt

If you are interested in the losses as well, they can be found in this notebook in the output windows.