forked from MFajcik/RumourEval2019
-
Notifications
You must be signed in to change notification settings - Fork 0
/
config.json
280 lines (280 loc) · 9.21 KB
/
config.json
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
{
"active_model": "BERT_textonly",
"models": {
"ensemble": {
"modeltype": "bert-large-uncased",
"hyperparameters": {
"batch_size": 2,
"max_length": 200,
"learning_rate": 8e-4,
"ensemble_models": 17
},
"train_data": "data_preprocessing/saved_data_RumEval2019_SEQ/train/train.json",
"dev_data": "data_preprocessing/saved_data_RumEval2019_SEQ/dev/dev.json",
"test_data": "data_preprocessing/saved_data_RumEval2019_TEST/test/test.json",
"cuda": true
},
"baseline": {
"hyperparameters": {
"learning_rate": 5e-6,
"inp_size": 320,
"dropout_rate": 0.5,
"batch_size": 16,
"epochs": 70,
"sep_token": "<sep>",
"FC_layers": 1,
"FC_size": 320
},
"train_data": "data_preprocessing/saved_data_RumEval2019_BRANCH/train/train.json",
"dev_data": "data_preprocessing/saved_data_RumEval2019_BRANCH/dev/dev.json",
"cuda": true,
"embeddings": "fasttext.en.300d",
"vector_cache": "../Figurative_Language/.vector_cache"
},
"baseline_LSTM": {
"hyperparameters": {
"learning_rate": 0.00008,
"dropout_rate": 0.5,
"batch_size": 16,
"epochs": 70,
"sep_token": "<sep>",
"RNN_nhidden": 314,
"RNN_layers": 1,
"RNN_input_dim": 314,
"RNN_clip": 0.5,
"FC_layers": 2,
"FC_size": 320
},
"train_data": "data_preprocessing/saved_data_RumEval2019_BRANCH/train/train.json",
"dev_data": "data_preprocessing/saved_data_RumEval2019_BRANCH/dev/dev.json",
"cuda": true,
"embeddings": "fasttext.en.300d",
"vector_cache": "../Figurative_Language/.vector_cache"
},
"selfatt_textonly": {
"hyperparameters": {
"learning_rate": 1e-05,
"dropout_rate": 0.5,
"batch_size": 16,
"epochs": 70,
"sep_token": "<sep>",
"RNN_nhidden": 300,
"RNN_layers": 1,
"RNN_input_dim": 300,
"RNN_clip": 0.5,
"ATTENTION_nhidden": 320,
"ATTENTION_hops": 10,
"FC_layers": 0,
"FC_size": 300,
"cov_penalization": 0,
"optimize_embeddings": true
},
"train_data": "data_preprocessing/saved_data_RumEval2019_BRANCH/train/train.json",
"dev_data": "data_preprocessing/saved_data_RumEval2019_BRANCH/dev/dev.json",
"cuda": true,
"embeddings": "fasttext.en.300d",
"vector_cache": "../Figurative_Language/.vector_cache"
},
"selfatt_textonly_seq": {
"hyperparameters": {
"learning_rate": 5e-06,
"dropout_rate": 0.5,
"batch_size": 32,
"epochs": 70,
"sep_token": "<sep>",
"RNN_nhidden": 300,
"RNN_layers": 2,
"RNN_input_dim": 300,
"RNN_clip": 0.5,
"ATTENTION_nhidden": 320,
"ATTENTION_hops": 10,
"FC_layers": 0,
"FC_size": 300,
"cov_penalization": 0,
"scale_emb_grad_by_freq": true,
"optimize_embeddings": true
},
"train_data": "data_preprocessing/saved_data_RumEval2019_SEQ_ALL/train/train.json",
"dev_data": "data_preprocessing/saved_data_RumEval2019_SEQ_ALL/dev/dev.json",
"cuda": true,
"embeddings": "fasttext.en.300d",
"vector_cache": "../Figurative_Language/.vector_cache"
},
"self_att_with_bert_tokenizer": {
"variant": "bert-base-uncased",
"hyperparameters": {
"learning_rate": 1e-5,
"dropout_rate": 0.5,
"batch_size": 32,
"epochs": 200,
"RNN_nhidden": 384,
"RNN_layers": 1,
"RNN_input_dim": 768,
"RNN_clip": 0.5,
"ATTENTION_nhidden": 788,
"ATTENTION_hops": 10,
"FC_layers": 0,
"FC_size": 768,
"cov_penalization": 0,
"scale_emb_grad_by_freq": true,
"optimize_embeddings": true,
"max_length": 512,
"attention_probs_dropout_prob": 0.5,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.5,
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 3072,
"max_position_embeddings": 512,
"num_attention_heads": 12,
"num_hidden_layers": 12,
"type_vocab_size": 2,
"vocab_size": 30522
},
"train_data": "data_preprocessing/saved_data_RumEval2019_SEQ_ALL/train/train.json",
"dev_data": "data_preprocessing/saved_data_RumEval2019_SEQ_ALL/dev/dev.json",
"test_data": "data_preprocessing/saved_data_RumEval2019_TEST/test/test_labelled.json",
"cuda": true,
"embeddings": "fasttext.en.300d",
"vector_cache": "../Figurative_Language/.vector_cache"
},
"features_seq": {
"hyperparameters": {
"learning_rate": 0.0008,
"inp_size": 319,
"dropout_rate": 0.5,
"batch_size": 16,
"epochs": 120,
"sep_token": "<sep>",
"FC_layers": 2,
"FC_size": 319
},
"train_data": "data_preprocessing/saved_data_RumEval2019_SEQ_ALL/train/train.json",
"dev_data": "data_preprocessing/saved_data_RumEval2019_SEQ_ALL/dev/dev.json",
"test_data": "data_preprocessing/saved_data_RumEval2019_SEQ_ALL_TEST/test/test_labelled.json",
"cuda": true,
"embeddings": "fasttext.en.300d",
"vector_cache": "../Figurative_Language/.vector_cache"
},
"BERT_textonly": {
"early_stop_after": 4,
"variant": "bert-large-uncased",
"hyperparameters": {
"learning_rate": 1.5e-06,
"batch_size": 2,
"true_batch_size": 32,
"max_length": 200,
"epochs": 90,
"attention_probs_dropout_prob": 0.5,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 3072,
"max_position_embeddings": 512,
"num_attention_heads": 12,
"num_hidden_layers": 12,
"type_vocab_size": 2,
"vocab_size": 30522
},
"train_data": "data_preprocessing/saved_data_RumEval2019/train/train.json",
"dev_data": "data_preprocessing/saved_data_RumEval2019/dev/dev.json",
"test_data": "data_preprocessing/saved_data_RumEval2019/test/test.json",
"cuda": true
},
"BERT_introspection": {
"hyperparameters": {
"learning_rate": 1.5e-06,
"batch_size": 1,
"true_batch_size": 32,
"max_length": 200,
"epochs": 90,
"attention_probs_dropout_prob": 0.5,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 3072,
"max_position_embeddings": 512,
"num_attention_heads": 12,
"num_hidden_layers": 12,
"type_vocab_size": 2,
"vocab_size": 30522
},
"train_data": "data_preprocessing/saved_data_RumEval2019_SEQ/train/train.json",
"dev_data": "data_preprocessing/saved_data_RumEval2019_SEQ/dev/dev.json",
"cuda": true
},
"BERT_veracity": {
"hyperparameters": {
"learning_rate": 0.000005,
"batch_size": 8,
"true_batch_size": 24,
"max_length": 180,
"epochs": 90,
"attention_probs_dropout_prob": 0.4,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.4,
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 3072,
"max_position_embeddings": 512,
"num_attention_heads": 12,
"num_hidden_layers": 12,
"type_vocab_size": 2,
"vocab_size": 30522
},
"train_data": "data_preprocessing/saved_data_RumEval2019_Veracity/train/train.json",
"dev_data": "data_preprocessing/saved_data_RumEval2019_Veracity/dev/dev.json",
"cuda": true
},
"BERT_withf": {
"hyperparameters": {
"learning_rate": 0.000001,
"batch_size": 2,
"true_batch_size": 16,
"max_length": 256,
"epochs": 150,
"attention_probs_dropout_prob": 0.5,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.5,
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 3072,
"max_position_embeddings": 512,
"num_attention_heads": 12,
"num_hidden_layers": 12,
"type_vocab_size": 2,
"vocab_size": 30522
},
"train_data": "data_preprocessing/saved_data_RumEval2019_SEQ_ALL/train/train.json",
"dev_data": "data_preprocessing/saved_data_RumEval2019_SEQ_ALL/dev/dev.json",
"cuda": true
},
"selfatt_text_and_baseline": {
"hyperparameters": {
"learning_rate": 0.00001,
"inp_size": 314,
"dropout_rate": 0.5,
"batch_size": 16,
"epochs": 500,
"sep_token": "<sep>",
"RNN_nhidden": 300,
"RNN_layers": 1,
"RNN_input_dim": 300,
"RNN_clip": 0.5,
"ATTENTION_nhidden": 320,
"ATTENTION_hops": 10,
"FC_layers": 2,
"FC_size": 314,
"cov_penalization": 0.0,
"optimize_embeddings": true
},
"train_data": "data_preprocessing/saved_data_RumEval2019/train/train.json",
"dev_data": "data_preprocessing/saved_data_RumEval2019/dev/dev.json",
"cuda": true,
"embeddings": "fasttext.en.300d",
"vector_cache": "../Figurative_Language/.vector_cache"
}
}
}