-
Notifications
You must be signed in to change notification settings - Fork 0
/
software.py
288 lines (249 loc) · 12.1 KB
/
software.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
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
import os
import sys
import datetime
import pandas as pd
import torch
import torch.nn as nn
import copy
sys.path.append('../')
from lr_scheduler import CyclicLR
from training_utils import training_loop, test_loop
from model import RNNLM
from data_utils import (IndexVectorizer,
TextDataset,
SpacyTokenizer,
LMDataLoader)
import pickle
class SMaPPLearn:
def __init__(self, data_dir, train_file, valid_file,
max_vocab_size = 20000, batch_size = 50, revectorize = False):
self.data_dir = data_dir
self.TOKENIZE = SpacyTokenizer().tokenize
self.MIN_WORD_FREQ = 2
self.MAX_VOCAB_SIZE = max_vocab_size
self.STAT_END_TOK = True
## Training Language Model
self.batch_size = batch_size
self.target_seq_len = 65
self.max_seq_len = 75
self.min_seq_len = 5
# GPU setup
self.use_gpu = torch.cuda.is_available()
device_num = 0
self.device = torch.device(f"cuda:{device_num}" if self.use_gpu
else "cpu")
# IO setup
today = datetime.datetime.now().strftime('%Y-%m-%d')
model_cache_dir = os.path.join(data_dir, 'models')
self.data_cache = os.path.join(model_cache_dir, 'data_cache.pkl')
self.vectorizer_cache = os.path.join(model_cache_dir,
'lm_vectorizer.pkl')
os.makedirs(model_cache_dir, exist_ok=True)
self.model_file_lm = os.path.join(model_cache_dir, f'LM__{today}.json')
self.train_file = train_file
self.valid_file = valid_file
self.revectorize = revectorize
if self.revectorize or not os.path.isfile(self.data_cache):
print("Vectorizing starting...")
train = pd.read_csv(self.train_file)
valid = pd.read_csv(self.valid_file)
self.vectorizer = IndexVectorizer(max_words = self.MAX_VOCAB_SIZE,
min_frequency=self.MIN_WORD_FREQ,
start_end_tokens=self.STAT_END_TOK,
tokenize=self.TOKENIZE)
self.train_ds = TextDataset(data=train, vectorizer=self.vectorizer,
text_col='text')
self.valid_ds = TextDataset(data=valid, vectorizer=self.vectorizer,
text_col='text')
pickle.dump([self.train_ds, self.valid_ds], open(self.data_cache,
'wb'))
pickle.dump(self.vectorizer, open(self.vectorizer_cache, 'wb'))
else:
self.train_ds, self.valid_ds = pickle.load(open(self.data_cache,
'rb'))
self.vectorizer = pickle.load(open(self.vectorizer_cache, 'rb'))
print("Vectorizing is complete.")
print(f'Train size: {len(self.train_ds)}\n \
valid size:{len(self.valid_ds)}')
print(f"Vocab size: {len(self.vectorizer.vocabulary)}")
self.train_dl = LMDataLoader(dataset=self.train_ds,
target_seq_len=self.target_seq_len,
shuffle=True,
max_seq_len=self.max_seq_len,
min_seq_len=self.min_seq_len,
p_half_seq_len=0.05,
batch_size=self.batch_size)
self.valid_dl = LMDataLoader(dataset=self.valid_ds,
target_seq_len=self.target_seq_len,
shuffle=True,
max_seq_len=self.max_seq_len,
min_seq_len=self.min_seq_len,
p_half_seq_len=0.05,
batch_size=self.batch_size)
print("Created Data Loaders for documents")
def freezeTo(self, n):
if n>1 and n<=1+self.lstm_layers:
n = 1 + (n-1)*4
print("Freezing layers until ", n)
i=0
for name, p in self.lm.named_parameters():
if i<n or n==-1:
p.requires_grad = False
else:
p.requires_grad = True
i+=1
for name, p in self.lm.named_parameters():
print(name, p.requires_grad)
def fit_language_model(self, pretrained_itos = None,
pretrained_weight_file = None, lm_hidden_dim = 1150,
lm_embedding_dim = 400, lm_lstm_layers = 3, num_epochs=100,
display_epoch_freq = 1, scheduler = 'ulmfit',
max_lrs = [1e-3, 1e-3, 1e-3, 1e-3, 1e-3]):
## Model Architecture
self.hidden_dim = lm_hidden_dim
self.embedding_dim = lm_embedding_dim
self.dropout = 0.3
self.lstm_layers = lm_lstm_layers
self.lstm_bidirection = False
self.lstm_tie_weights = True
self.num_epochs = num_epochs
self.display_epoch_freq = display_epoch_freq
if self.use_gpu: torch.cuda.manual_seed(303)
else: torch.manual_seed(303)
# set up Files to save stuff in
runtime = datetime.datetime.now().strftime('%Y-%m-%d_%H-%M-%S')
if pretrained_weight_file is not None and pretrained_itos is not None:
print("Starting Loading pretrained Wikitext model")
enc = torch.load(pretrained_weight_file,
map_location=lambda storage, loc: storage)
self.embedding_dim = enc['0.encoder.weight'].shape[1]
self.hidden_dim = int(
enc['0.rnns.0.module.weight_hh_l0_raw'].shape[0]/4)
self.lstm_layers = 3
new_enc = {}
for k,v in enc.items():
layer_detail = k.split('.')
layer_name = layer_detail[-1].replace('_raw', '')
if len(layer_detail) == self.lstm_layers:
new_enc[f'{layer_detail[1]}.{layer_name}'] = v
else:
new_enc[f'{layer_detail[1]}.{layer_detail[2]}.{layer_name}'
] = v
del new_enc['encoder_with_dropout.embed.weight']
pretrained_idx2word = pickle.load(open(pretrained_itos, 'rb'))
pretrained_word2idx =\
{k: i for i,k in enumerate(pretrained_idx2word)}
new_model_vectorizer = self.vectorizer
pretrained_encoder_weights = enc['0.encoder.weight']
row_m = pretrained_encoder_weights.mean(dim=0)
row_m = [x.item() for x in row_m]
new_vocab_size = len(new_model_vectorizer.word2idx)
new_encoder_weights = torch.tensor(
[row_m for i in range(new_vocab_size)])
new_idx2weights = {}
for word, i in new_model_vectorizer.word2idx.items():
if word in pretrained_word2idx:
word_idx = pretrained_word2idx[word]
new_encoder_weights[i] =\
pretrained_encoder_weights[word_idx]
new_enc['encoder.weight'] = new_encoder_weights
new_enc['decoder.weight'] = copy.copy(new_encoder_weights)
new_enc['decoder.bias'] =\
torch.zeros(new_enc['decoder.weight'].shape[0])
self.lm = RNNLM(device=self.device, vocab_size=new_vocab_size,
embedding_size=self.embedding_dim, hidden_size=self.hidden_dim,
batch_size=50, num_layers=3, tie_weights=True,
word2idx = new_model_vectorizer.word2idx)
print("Initialised loading with pretrained Wikitext model")
else:
# Build and initialize the model
print("No Pretrained Model specified")
self.lm = RNNLM(self.device, self.vectorizer.vocabulary_size,
self.embedding_dim, self.hidden_dim,
self.batch_size, dropout = self.dropout,
tie_weights = self.lstm_tie_weights,
num_layers = self.lstm_layers,
bidirectional = self.lstm_bidirection,
word2idx = self.vectorizer.word2idx,
log_softmax = False)
if self.use_gpu:
self.lm = self.lm.to(self.device)
# Loss and Optimizer
self.loss = nn.CrossEntropyLoss()
# Extract pointers to the parameters of the lstms
param_list = [{'params': rnn.parameters(), 'lr': 1e-3}
for rnn in self.lm.rnns]
# If weights are tied between encoder and decoder, we can only optimize
# parameters in one of those two layers
if not self.lstm_tie_weights:
param_list.extend([
{'params': self.lm.encoder.parameters(), 'lr':1e-3},
{'params': self.lm.decoder.parameters(), 'lr':1e-3},
])
else:
param_list.extend([
{'params': self.lm.decoder.parameters(), 'lr':1e-3},
])
self.optimizer = torch.optim.Adam(param_list)
if scheduler == 'ulmfit':
self.scheduler = CyclicLR(self.optimizer, max_lrs=max_lrs,
mode='ulmfit', ratio=1.5, cut_frac=0.4,
train_data_loader = self.train_dl,
verbose=False)
print("Beginning LM Fine Tuning")
self.freezeTo(3)
history = training_loop(batch_size=self.batch_size,
num_epochs=1,
display_freq=self.display_epoch_freq,
model=self.lm,
criterion=self.loss,
optim=self.optimizer,
scheduler=self.scheduler,
device=self.device,
training_set=self.train_dl,
validation_set=self.valid_dl,
best_model_path=self.model_file_lm,
history=None)
self.freezeTo(2)
history = training_loop(batch_size=self.batch_size,
num_epochs=1,
display_freq=self.display_epoch_freq,
model=self.lm,
criterion=self.loss,
optim=self.optimizer,
scheduler=self.scheduler,
device=self.device,
training_set=self.train_dl,
validation_set=self.valid_dl,
best_model_path=self.model_file_lm,
history = history)
self.freezeTo(0)
history = training_loop(batch_size=self.batch_size,
num_epochs=self.num_epochs-2,
display_freq=self.display_epoch_freq,
model=self.lm,
criterion=self.loss,
optim=self.optimizer,
scheduler=self.scheduler,
device=self.device,
training_set=self.train_dl,
validation_set=self.valid_dl,
best_model_path=self.model_file_lm,
history = history)
if __name__ == '__main__':
test = SMaPPLearn(data_dir = '../data/imdb/',
train_file = '../data/imdb/unsup.csv',
valid_file = '../data/imdb/valid.csv',
max_vocab_size = 20000, revectorize = False)
"""
test.fit_language_model(lm_embedding_dim = 200, lm_hidden_dim = 250,
num_epochs = 10, display_epoch_freq = 1)
"""
test.fit_language_model(
pretrained_weight_file =
'../data/imdb/weights_pretrained/fwd_wt103.h5',
pretrained_itos =
'../data/imdb/weights_pretrained/fitos_wt103.pkl',
display_epoch_freq = 1, num_epochs = 15, scheduler = 'ulmfit',
max_lrs = [1e-3, 1e-3, 1e-3, 1e-3, 1e-3])
#"""