-
Notifications
You must be signed in to change notification settings - Fork 0
/
try.py
46 lines (33 loc) · 1.04 KB
/
try.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
import json
from keras.preprocessing.text import Tokenizer,one_hot, text_to_word_sequence
from unidecode import unidecode
import cPickle as pickle
import numpy as np
from keras.models import Model
from model import PredictionDecoderModel
"""print "Loading data"
with open("data/preprocessed_data.pkl") as f:
data = pickle.load(f)
print "Done"
reclen = 800
context = np.asarray(data["context"][:reclen])
qn_output = np.asarray(data["qn_output"][:reclen])
answer = np.asarray(data["answer"][:reclen])
qn_input = np.asarray(data["qn_input"][:reclen])
data = {
"context": context,
"qn_output": qn_output,
"qn_input": qn_input,
"answer": answer
}
with open("data/preprocessed_data_trimmed.pkl", "wb") as f:
pickle.dump(data, f, protocol=pickle.HIGHEST_PROTOCOL)"""
"""import json
print('Reading SQuAD data... ')
with open("data/train_parsed_trimmed.pkl") as fd:
samples = pickle.load(f)
print('Done!')
print len(samples)
model = TrainingModel()
model.load_weights("pre-trained/current.hdf5")"""
model = PredictionDecoderModel()