Hugging Face's logo

Stevenf232
/
hypencoder_2layer_SapBERT_nameOnly

PyTorch
Model card Files
xet
 Community
Stevenf232 commited on
Commit
4869432
·
verified ·
1 Parent(s): b81796c

Upload trained model from Colab

Browse files
Files changed (12) hide show
  1. config.json +40 -0
  2. optimizer.pt +3 -0
  3. pytorch_model.bin +3 -0
  4. rng_state.pth +3 -0
  5. scaler.pt +3 -0
  6. scheduler.pt +3 -0
  7. special_tokens_map.json +7 -0
  8. tokenizer.json +0 -0
  9. tokenizer_config.json +59 -0
  10. trainer_state.json +334 -0
  11. training_args.bin +3 -0
  12. vocab.txt +0 -0
config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "HypencoderDualEncoder"
4
+ ],
5
+ "loss_kwargs": [
6
+ {
7
+ "only_use_first_item": false,
8
+ "use_in_batch_negatives": true
9
+ }
10
+ ],
11
+ "loss_type": [
12
+ "cross_entropy"
13
+ ],
14
+ "passage_encoder_kwargs": {
15
+ "freeze_transformer": false,
16
+ "model_name_or_path": "cambridgeltl/SapBERT-from-PubMedBERT-fulltext",
17
+ "pooling_type": "cls"
18
+ },
19
+ "passage_encoder_type": "",
20
+ "query_encoder_kwargs": {
21
+ "base_encoder_output_dim": 768,
22
+ "converter_kwargs": {
23
+ "activation_type": "relu",
24
+ "do_residual_on_last": false,
25
+ "vector_dimensions": [
26
+ 768,
27
+ 768,
28
+ 768,
29
+ 1
30
+ ]
31
+ },
32
+ "embedding_representation": null,
33
+ "freeze_transformer": false,
34
+ "model_name_or_path": "cambridgeltl/SapBERT-from-PubMedBERT-fulltext"
35
+ },
36
+ "query_encoder_type": "",
37
+ "shared_encoder": true,
38
+ "torch_dtype": "float32",
39
+ "transformers_version": "4.50.0"
40
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:21ea3134ff8026463a1d5e6aeab1906712da6bf90d53ee53b2205d9483831a3f
3
+ size 961114315
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41312181fe5b44a27f26bcaa0918bc70ac04fd20fd2417ea908dca1ec675a0e5
3
+ size 482938527
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:120f55fef9769a04b7594b47a1c65d897aa1729ea24eed7073ff72ca782ac5e2
3
+ size 14645
scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f31bcd9e60dde16db96b3b8cfe15dac15e8e1911393a58739b14a4c0a0ea9ed
3
+ size 1383
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ce07953d081622886b64e5807e3eebaab5f992052937997d3feb31d45c36853
3
+ size 1465
special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "extra_special_tokens": {},
49
+ "full_tokenizer_file": null,
50
+ "mask_token": "[MASK]",
51
+ "model_max_length": 1000000000000000019884624838656,
52
+ "never_split": null,
53
+ "pad_token": "[PAD]",
54
+ "sep_token": "[SEP]",
55
+ "strip_accents": null,
56
+ "tokenize_chinese_chars": true,
57
+ "tokenizer_class": "BertTokenizer",
58
+ "unk_token": "[UNK]"
59
+ }
trainer_state.json ADDED
@@ -0,0 +1,334 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 4.761904761904762,
6
+ "eval_steps": 10,
7
+ "global_step": 200,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.23809523809523808,
14
+ "grad_norm": 107.2943344116211,
15
+ "learning_rate": 7e-06,
16
+ "loss": 4.5568,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.23809523809523808,
21
+ "eval_loss": 4.329397678375244,
22
+ "eval_runtime": 2.6051,
23
+ "eval_samples_per_second": 982.322,
24
+ "eval_steps_per_second": 15.355,
25
+ "step": 10
26
+ },
27
+ {
28
+ "epoch": 0.47619047619047616,
29
+ "grad_norm": 82.67948150634766,
30
+ "learning_rate": 1.7e-05,
31
+ "loss": 3.937,
32
+ "step": 20
33
+ },
34
+ {
35
+ "epoch": 0.47619047619047616,
36
+ "eval_loss": 3.166707754135132,
37
+ "eval_runtime": 2.6689,
38
+ "eval_samples_per_second": 958.832,
39
+ "eval_steps_per_second": 14.988,
40
+ "step": 20
41
+ },
42
+ {
43
+ "epoch": 0.7142857142857143,
44
+ "grad_norm": 81.49203491210938,
45
+ "learning_rate": 1.9222222222222225e-05,
46
+ "loss": 2.7135,
47
+ "step": 30
48
+ },
49
+ {
50
+ "epoch": 0.7142857142857143,
51
+ "eval_loss": 1.7821017503738403,
52
+ "eval_runtime": 3.0515,
53
+ "eval_samples_per_second": 838.597,
54
+ "eval_steps_per_second": 13.108,
55
+ "step": 30
56
+ },
57
+ {
58
+ "epoch": 0.9523809523809523,
59
+ "grad_norm": 67.98638153076172,
60
+ "learning_rate": 1.8111111111111112e-05,
61
+ "loss": 1.6075,
62
+ "step": 40
63
+ },
64
+ {
65
+ "epoch": 0.9523809523809523,
66
+ "eval_loss": 0.9218789339065552,
67
+ "eval_runtime": 2.6581,
68
+ "eval_samples_per_second": 962.702,
69
+ "eval_steps_per_second": 15.048,
70
+ "step": 40
71
+ },
72
+ {
73
+ "epoch": 1.1904761904761905,
74
+ "grad_norm": 41.94712448120117,
75
+ "learning_rate": 1.7e-05,
76
+ "loss": 0.7666,
77
+ "step": 50
78
+ },
79
+ {
80
+ "epoch": 1.1904761904761905,
81
+ "eval_loss": 0.5395554304122925,
82
+ "eval_runtime": 2.7233,
83
+ "eval_samples_per_second": 939.664,
84
+ "eval_steps_per_second": 14.688,
85
+ "step": 50
86
+ },
87
+ {
88
+ "epoch": 1.4285714285714286,
89
+ "grad_norm": 45.76442337036133,
90
+ "learning_rate": 1.588888888888889e-05,
91
+ "loss": 0.4077,
92
+ "step": 60
93
+ },
94
+ {
95
+ "epoch": 1.4285714285714286,
96
+ "eval_loss": 0.395812064409256,
97
+ "eval_runtime": 2.6568,
98
+ "eval_samples_per_second": 963.196,
99
+ "eval_steps_per_second": 15.056,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 1.6666666666666665,
104
+ "grad_norm": 36.85358428955078,
105
+ "learning_rate": 1.477777777777778e-05,
106
+ "loss": 0.2843,
107
+ "step": 70
108
+ },
109
+ {
110
+ "epoch": 1.6666666666666665,
111
+ "eval_loss": 0.3538413345813751,
112
+ "eval_runtime": 2.958,
113
+ "eval_samples_per_second": 865.099,
114
+ "eval_steps_per_second": 13.522,
115
+ "step": 70
116
+ },
117
+ {
118
+ "epoch": 1.9047619047619047,
119
+ "grad_norm": 17.167036056518555,
120
+ "learning_rate": 1.3666666666666667e-05,
121
+ "loss": 0.2165,
122
+ "step": 80
123
+ },
124
+ {
125
+ "epoch": 1.9047619047619047,
126
+ "eval_loss": 0.3396788537502289,
127
+ "eval_runtime": 3.0374,
128
+ "eval_samples_per_second": 842.51,
129
+ "eval_steps_per_second": 13.169,
130
+ "step": 80
131
+ },
132
+ {
133
+ "epoch": 2.142857142857143,
134
+ "grad_norm": 28.207691192626953,
135
+ "learning_rate": 1.2555555555555557e-05,
136
+ "loss": 0.1979,
137
+ "step": 90
138
+ },
139
+ {
140
+ "epoch": 2.142857142857143,
141
+ "eval_loss": 0.32209891080856323,
142
+ "eval_runtime": 2.6763,
143
+ "eval_samples_per_second": 956.167,
144
+ "eval_steps_per_second": 14.946,
145
+ "step": 90
146
+ },
147
+ {
148
+ "epoch": 2.380952380952381,
149
+ "grad_norm": 20.979476928710938,
150
+ "learning_rate": 1.1444444444444444e-05,
151
+ "loss": 0.1476,
152
+ "step": 100
153
+ },
154
+ {
155
+ "epoch": 2.380952380952381,
156
+ "eval_loss": 0.30111950635910034,
157
+ "eval_runtime": 2.7597,
158
+ "eval_samples_per_second": 927.283,
159
+ "eval_steps_per_second": 14.494,
160
+ "step": 100
161
+ },
162
+ {
163
+ "epoch": 2.619047619047619,
164
+ "grad_norm": 14.82142448425293,
165
+ "learning_rate": 1.0333333333333335e-05,
166
+ "loss": 0.1061,
167
+ "step": 110
168
+ },
169
+ {
170
+ "epoch": 2.619047619047619,
171
+ "eval_loss": 0.30764198303222656,
172
+ "eval_runtime": 2.701,
173
+ "eval_samples_per_second": 947.422,
174
+ "eval_steps_per_second": 14.809,
175
+ "step": 110
176
+ },
177
+ {
178
+ "epoch": 2.857142857142857,
179
+ "grad_norm": 19.38449478149414,
180
+ "learning_rate": 9.222222222222224e-06,
181
+ "loss": 0.0907,
182
+ "step": 120
183
+ },
184
+ {
185
+ "epoch": 2.857142857142857,
186
+ "eval_loss": 0.31123870611190796,
187
+ "eval_runtime": 2.7275,
188
+ "eval_samples_per_second": 938.22,
189
+ "eval_steps_per_second": 14.665,
190
+ "step": 120
191
+ },
192
+ {
193
+ "epoch": 3.0952380952380953,
194
+ "grad_norm": 15.523360252380371,
195
+ "learning_rate": 8.111111111111112e-06,
196
+ "loss": 0.0573,
197
+ "step": 130
198
+ },
199
+ {
200
+ "epoch": 3.0952380952380953,
201
+ "eval_loss": 0.306456595659256,
202
+ "eval_runtime": 3.037,
203
+ "eval_samples_per_second": 842.596,
204
+ "eval_steps_per_second": 13.171,
205
+ "step": 130
206
+ },
207
+ {
208
+ "epoch": 3.3333333333333335,
209
+ "grad_norm": 9.912215232849121,
210
+ "learning_rate": 7e-06,
211
+ "loss": 0.0638,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 3.3333333333333335,
216
+ "eval_loss": 0.3030768036842346,
217
+ "eval_runtime": 2.784,
218
+ "eval_samples_per_second": 919.181,
219
+ "eval_steps_per_second": 14.368,
220
+ "step": 140
221
+ },
222
+ {
223
+ "epoch": 3.571428571428571,
224
+ "grad_norm": 32.101985931396484,
225
+ "learning_rate": 5.88888888888889e-06,
226
+ "loss": 0.0716,
227
+ "step": 150
228
+ },
229
+ {
230
+ "epoch": 3.571428571428571,
231
+ "eval_loss": 0.2990337312221527,
232
+ "eval_runtime": 2.8128,
233
+ "eval_samples_per_second": 909.761,
234
+ "eval_steps_per_second": 14.221,
235
+ "step": 150
236
+ },
237
+ {
238
+ "epoch": 3.8095238095238093,
239
+ "grad_norm": 17.976055145263672,
240
+ "learning_rate": 4.777777777777778e-06,
241
+ "loss": 0.0841,
242
+ "step": 160
243
+ },
244
+ {
245
+ "epoch": 3.8095238095238093,
246
+ "eval_loss": 0.3003748655319214,
247
+ "eval_runtime": 2.7098,
248
+ "eval_samples_per_second": 944.346,
249
+ "eval_steps_per_second": 14.761,
250
+ "step": 160
251
+ },
252
+ {
253
+ "epoch": 4.0476190476190474,
254
+ "grad_norm": 16.045225143432617,
255
+ "learning_rate": 3.6666666666666666e-06,
256
+ "loss": 0.096,
257
+ "step": 170
258
+ },
259
+ {
260
+ "epoch": 4.0476190476190474,
261
+ "eval_loss": 0.3065829873085022,
262
+ "eval_runtime": 2.7428,
263
+ "eval_samples_per_second": 933.004,
264
+ "eval_steps_per_second": 14.584,
265
+ "step": 170
266
+ },
267
+ {
268
+ "epoch": 4.285714285714286,
269
+ "grad_norm": 24.23316764831543,
270
+ "learning_rate": 2.5555555555555557e-06,
271
+ "loss": 0.0728,
272
+ "step": 180
273
+ },
274
+ {
275
+ "epoch": 4.285714285714286,
276
+ "eval_loss": 0.3128698170185089,
277
+ "eval_runtime": 3.0252,
278
+ "eval_samples_per_second": 845.905,
279
+ "eval_steps_per_second": 13.222,
280
+ "step": 180
281
+ },
282
+ {
283
+ "epoch": 4.523809523809524,
284
+ "grad_norm": 10.853073120117188,
285
+ "learning_rate": 1.4444444444444445e-06,
286
+ "loss": 0.0594,
287
+ "step": 190
288
+ },
289
+ {
290
+ "epoch": 4.523809523809524,
291
+ "eval_loss": 0.3163248300552368,
292
+ "eval_runtime": 2.763,
293
+ "eval_samples_per_second": 926.153,
294
+ "eval_steps_per_second": 14.477,
295
+ "step": 190
296
+ },
297
+ {
298
+ "epoch": 4.761904761904762,
299
+ "grad_norm": 18.616168975830078,
300
+ "learning_rate": 3.3333333333333335e-07,
301
+ "loss": 0.071,
302
+ "step": 200
303
+ },
304
+ {
305
+ "epoch": 4.761904761904762,
306
+ "eval_loss": 0.3181891143321991,
307
+ "eval_runtime": 3.0801,
308
+ "eval_samples_per_second": 830.819,
309
+ "eval_steps_per_second": 12.987,
310
+ "step": 200
311
+ }
312
+ ],
313
+ "logging_steps": 10,
314
+ "max_steps": 200,
315
+ "num_input_tokens_seen": 0,
316
+ "num_train_epochs": 5,
317
+ "save_steps": 2500,
318
+ "stateful_callbacks": {
319
+ "TrainerControl": {
320
+ "args": {
321
+ "should_epoch_stop": false,
322
+ "should_evaluate": false,
323
+ "should_log": false,
324
+ "should_save": true,
325
+ "should_training_stop": true
326
+ },
327
+ "attributes": {}
328
+ }
329
+ },
330
+ "total_flos": 0.0,
331
+ "train_batch_size": 64,
332
+ "trial_name": null,
333
+ "trial_params": null
334
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61af473bd8eb1d8960414eade93748ff6750021ee307cbdaee3c5db934ca4afe
3
+ size 5713
vocab.txt ADDED
The diff for this file is too large to render. See raw diff