-
Notifications
You must be signed in to change notification settings - Fork 1
/
evaluate.py
152 lines (131 loc) · 5.34 KB
/
evaluate.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
from __future__ import unicode_literals, print_function, division
from io import open
import unicodedata
import string
import re
import random
import pickle
import pandas as pd
import torch
import torch.nn as nn
from torch import optim
import torch.nn.functional as F
from config import MAX_LENGTH, SOS_token, EOS_token, device
f_read = open('input_lang_word2index.pkl', 'rb')
input_lang_word2index = pickle.load(f_read)
f_read.close()
f_read = open('output_lang_index2word.pkl', 'rb')
output_lang_index2word = pickle.load(f_read)
f_read.close()
def get_pairs(src_path, tgt_path):
src_lines = open(src_path, encoding='utf-8').read().strip().split('\n')
tgt_lines = open(tgt_path, encoding='utf-8').read().strip().split('\n')
pairs = []
for i in range(len(src_lines)):
lines = [src_lines[i], tgt_lines[i]]
pairs.append(lines)
return pairs
# def indexesFromSentence(sentence):
# indexes = []
# for word in sentence.split(' '):
# if word in input_lang_word2index:
# index = input_lang_word2index[word]
# indexes.append(index)
# else:
# indexes.append(3) #UNK
# return indexes
def indexesFromSentence(sentence):
return [input_lang_word2index[word] for word in sentence.split(' ')]
def tensorFromSentence(sentence):
indexes = indexesFromSentence(sentence)
indexes.append(EOS_token)
return torch.tensor(indexes, dtype=torch.long, device=device).view(-1, 1)
def evaluate(encoder, decoder, sentence, max_length=MAX_LENGTH):
with torch.no_grad():
input_tensor = tensorFromSentence(sentence)
input_length = input_tensor.size()[0]
encoder_hidden = encoder.initHidden()
encoder_outputs = torch.zeros(max_length, encoder.hidden_size, device=device)
for ei in range(input_length):
encoder_output, encoder_hidden = encoder(input_tensor[ei], encoder_hidden)
encoder_outputs[ei] += encoder_output[0, 0]
decoder_input = torch.tensor([[SOS_token]], device=device) # SOS
decoder_hidden = encoder_hidden
decoded_words = []
decoder_attentions = torch.zeros(max_length, max_length)
for di in range(max_length):
decoder_output, decoder_hidden, decoder_attention = decoder(decoder_input, decoder_hidden, encoder_outputs)
decoder_attentions[di] = decoder_attention.data
topv, topi = decoder_output.data.topk(1)
if topi.item() == EOS_token:
decoded_words.append('<EOS>')
break
else:
decoded_words.append(output_lang_index2word[topi.item()])
decoder_input = topi.squeeze().detach()
return decoded_words, decoder_attentions[:di + 1]
import rdkit
from rdkit import Chem
def can_smi(smi):
try:
mol = Chem.MolFromSmiles(smi)
smi = Chem.MolToSmiles(mol)
return smi
except:
print("output smiles is %s unvaild" % smi)
return None
def evaluateRandomly(encoder, decoder, pairs, n=10):
for i in range(n):
pair = random.choice(pairs)
print('>', pair[0].replace(" ", ""))
print('=', pair[1].replace(" ", ""))
output_words, attentions = evaluate(encoder, decoder, pair[0])
output_sentence = ''.join(output_words).replace("<EOS>", "")
output_sentence = can_smi(output_sentence)
print('<', output_sentence)
print('')
def evaluateAll(encoder, decoder, pairs):
top1 = 0
inputs = []
predictions = []
targets = []
matches = []
for pair in pairs:
try:
# print('>', pair[0].replace(" ", ""))
# print('=', pair[1].replace(" ", ""))
# print(pairs[0])
# if "%" in pair[0].replace(" ", "") or "%" in pair[1].replace(" ", ""):
# continue
# if 4 in pair[0].replace(" ", "") or 4 in pair[1].replace(" ", ""):
# continue
output_words, attentions = evaluate(encoder, decoder, pair[0])
inputs.append(pair[0].replace(" ", ""))
# print("output_words ", output_words)
output_sentence = ''.join(output_words).replace("<EOS>", "")
predictions.append(output_sentence)
targets.append(pair[1].replace(" ", ""))
output_sentence = can_smi(output_sentence)
# print('<', output_sentence)
# print('')
if output_sentence == can_smi(pair[1].replace(" ", "")):
print("正确", output_sentence)
top1 += 1
matches.append(1)
else:
matches.append(0)
except Exception as E:
print("APPEAR ERROR TOKEN", E)
continue
acc = top1/len(pairs) * 100
print("the final acc is ", acc, " %")
result_df = pd.DataFrame({"inputs": inputs, "targets": targets, "predictions": predictions, "matches":matches})
result_df.to_csv("预测结果.csv")
return acc
if __name__ == '__main__':
encoder1 = torch.load('model/AGEs_cml.cel/encoder.pth').to(device)
attn_decoder1 = torch.load('model/AGEs_cml.cel/decoder.pth').to(device)
val_pairs = get_pairs('data/test/src-test.txt', 'data/test/tgt-test.txt')
# val_pairs = get_pairs('data/AGEs_CML/src-val.txt', 'data/AGEs_CML/tgt-val.txt')
# evaluateRandomly(encoder1, attn_decoder1, pairs, n=1)
evaluateAll(encoder1, attn_decoder1, val_pairs)