Hugging Face's logo

gokulsrinivasagan
/
bert_tiny_lda_5_v1_book_stsb

Text Classification
Transformers
TensorBoard
Safetensors
English
distilbert
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
gokulsrinivasagan commited on
Commit
5142949
·
verified ·
1 Parent(s): 060eac6

End of training

Browse files
Files changed (6) hide show
  1. README.md +21 -6
  2. all_results.json +17 -0
  3. eval_results.json +11 -0
  4. logs/events.out.tfevents.1733838440.ki-g0008.683966.13 +3 -0
  5. train_results.json +9 -0
  6. trainer_state.json +429 -0
README.md CHANGED
@@ -1,13 +1,28 @@
1
  ---
2
  library_name: transformers
 
 
3
  base_model: gokulsrinivasagan/bert_tiny_lda_5_v1_book
4
  tags:
5
  - generated_from_trainer
 
 
6
  metrics:
7
  - spearmanr
8
  model-index:
9
  - name: bert_tiny_lda_5_v1_book_stsb
10
- results: []
 
 
 
 
 
 
 
 
 
 
 
11
  ---
12
 
13
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,12 +30,12 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # bert_tiny_lda_5_v1_book_stsb
17
 
18
- This model is a fine-tuned version of [gokulsrinivasagan/bert_tiny_lda_5_v1_book](https://huggingface.co/gokulsrinivasagan/bert_tiny_lda_5_v1_book) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 0.9757
21
- - Pearson: 0.7968
22
- - Spearmanr: 0.7942
23
- - Combined Score: 0.7955
24
 
25
  ## Model description
26
 
 
1
  ---
2
  library_name: transformers
3
+ language:
4
+ - en
5
  base_model: gokulsrinivasagan/bert_tiny_lda_5_v1_book
6
  tags:
7
  - generated_from_trainer
8
+ datasets:
9
+ - glue
10
  metrics:
11
  - spearmanr
12
  model-index:
13
  - name: bert_tiny_lda_5_v1_book_stsb
14
+ results:
15
+ - task:
16
+ name: Text Classification
17
+ type: text-classification
18
+ dataset:
19
+ name: GLUE STSB
20
+ type: glue
21
+ args: stsb
22
+ metrics:
23
+ - name: Spearmanr
24
+ type: spearmanr
25
+ value: 0.798781702305515
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
30
 
31
  # bert_tiny_lda_5_v1_book_stsb
32
 
33
+ This model is a fine-tuned version of [gokulsrinivasagan/bert_tiny_lda_5_v1_book](https://huggingface.co/gokulsrinivasagan/bert_tiny_lda_5_v1_book) on the GLUE STSB dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 0.8189
36
+ - Pearson: 0.8027
37
+ - Spearmanr: 0.7988
38
+ - Combined Score: 0.8007
39
 
40
  ## Model description
41
 
all_results.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 21.0,
3
+ "eval_combined_score": 0.8007289150909647,
4
+ "eval_loss": 0.8188873529434204,
5
+ "eval_pearson": 0.8026761278764143,
6
+ "eval_runtime": 0.4831,
7
+ "eval_samples": 1500,
8
+ "eval_samples_per_second": 3105.194,
9
+ "eval_spearmanr": 0.798781702305515,
10
+ "eval_steps_per_second": 12.421,
11
+ "total_flos": 3165851231294976.0,
12
+ "train_loss": 0.5645940970189823,
13
+ "train_runtime": 97.1506,
14
+ "train_samples": 5749,
15
+ "train_samples_per_second": 2958.809,
16
+ "train_steps_per_second": 11.837
17
+ }
eval_results.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 21.0,
3
+ "eval_combined_score": 0.8007289150909647,
4
+ "eval_loss": 0.8188873529434204,
5
+ "eval_pearson": 0.8026761278764143,
6
+ "eval_runtime": 0.4831,
7
+ "eval_samples": 1500,
8
+ "eval_samples_per_second": 3105.194,
9
+ "eval_spearmanr": 0.798781702305515,
10
+ "eval_steps_per_second": 12.421
11
+ }
logs/events.out.tfevents.1733838440.ki-g0008.683966.13 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4061dcf3dc74ee94615cbc4b92adc50fd929f8204e2ae1461ff06fe95abbdea
3
+ size 521
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 21.0,
3
+ "total_flos": 3165851231294976.0,
4
+ "train_loss": 0.5645940970189823,
5
+ "train_runtime": 97.1506,
6
+ "train_samples": 5749,
7
+ "train_samples_per_second": 2958.809,
8
+ "train_steps_per_second": 11.837
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,429 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8188873529434204,
3
+ "best_model_checkpoint": "bert_tiny_lda_5_v1_book_stsb/checkpoint-368",
4
+ "epoch": 21.0,
5
+ "eval_steps": 500,
6
+ "global_step": 483,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "grad_norm": 4.9694929122924805,
14
+ "learning_rate": 4.9e-05,
15
+ "loss": 2.8706,
16
+ "step": 23
17
+ },
18
+ {
19
+ "epoch": 1.0,
20
+ "eval_combined_score": 0.09954146094056157,
21
+ "eval_loss": 2.462472438812256,
22
+ "eval_pearson": 0.09907601559786483,
23
+ "eval_runtime": 0.4799,
24
+ "eval_samples_per_second": 3125.334,
25
+ "eval_spearmanr": 0.1000069062832583,
26
+ "eval_steps_per_second": 12.501,
27
+ "step": 23
28
+ },
29
+ {
30
+ "epoch": 2.0,
31
+ "grad_norm": 22.507034301757812,
32
+ "learning_rate": 4.8e-05,
33
+ "loss": 1.7766,
34
+ "step": 46
35
+ },
36
+ {
37
+ "epoch": 2.0,
38
+ "eval_combined_score": 0.6395870539114769,
39
+ "eval_loss": 1.3386552333831787,
40
+ "eval_pearson": 0.6420563993485077,
41
+ "eval_runtime": 0.5098,
42
+ "eval_samples_per_second": 2942.208,
43
+ "eval_spearmanr": 0.637117708474446,
44
+ "eval_steps_per_second": 11.769,
45
+ "step": 46
46
+ },
47
+ {
48
+ "epoch": 3.0,
49
+ "grad_norm": 20.13024139404297,
50
+ "learning_rate": 4.7e-05,
51
+ "loss": 1.1649,
52
+ "step": 69
53
+ },
54
+ {
55
+ "epoch": 3.0,
56
+ "eval_combined_score": 0.7499948802864915,
57
+ "eval_loss": 0.9883082509040833,
58
+ "eval_pearson": 0.7530220350451059,
59
+ "eval_runtime": 0.4998,
60
+ "eval_samples_per_second": 3001.186,
61
+ "eval_spearmanr": 0.746967725527877,
62
+ "eval_steps_per_second": 12.005,
63
+ "step": 69
64
+ },
65
+ {
66
+ "epoch": 4.0,
67
+ "grad_norm": 7.193154335021973,
68
+ "learning_rate": 4.600000000000001e-05,
69
+ "loss": 0.8446,
70
+ "step": 92
71
+ },
72
+ {
73
+ "epoch": 4.0,
74
+ "eval_combined_score": 0.7814449278644269,
75
+ "eval_loss": 0.9500728249549866,
76
+ "eval_pearson": 0.7801592206629702,
77
+ "eval_runtime": 0.4748,
78
+ "eval_samples_per_second": 3159.192,
79
+ "eval_spearmanr": 0.7827306350658837,
80
+ "eval_steps_per_second": 12.637,
81
+ "step": 92
82
+ },
83
+ {
84
+ "epoch": 5.0,
85
+ "grad_norm": 15.801517486572266,
86
+ "learning_rate": 4.5e-05,
87
+ "loss": 0.6587,
88
+ "step": 115
89
+ },
90
+ {
91
+ "epoch": 5.0,
92
+ "eval_combined_score": 0.7831183677075164,
93
+ "eval_loss": 0.9120289087295532,
94
+ "eval_pearson": 0.7838836623851876,
95
+ "eval_runtime": 0.4934,
96
+ "eval_samples_per_second": 3040.098,
97
+ "eval_spearmanr": 0.7823530730298451,
98
+ "eval_steps_per_second": 12.16,
99
+ "step": 115
100
+ },
101
+ {
102
+ "epoch": 6.0,
103
+ "grad_norm": 14.146763801574707,
104
+ "learning_rate": 4.4000000000000006e-05,
105
+ "loss": 0.5587,
106
+ "step": 138
107
+ },
108
+ {
109
+ "epoch": 6.0,
110
+ "eval_combined_score": 0.7878115431321093,
111
+ "eval_loss": 0.8917849659919739,
112
+ "eval_pearson": 0.7876118713697591,
113
+ "eval_runtime": 0.4812,
114
+ "eval_samples_per_second": 3117.386,
115
+ "eval_spearmanr": 0.7880112148944596,
116
+ "eval_steps_per_second": 12.47,
117
+ "step": 138
118
+ },
119
+ {
120
+ "epoch": 7.0,
121
+ "grad_norm": 12.302091598510742,
122
+ "learning_rate": 4.3e-05,
123
+ "loss": 0.5104,
124
+ "step": 161
125
+ },
126
+ {
127
+ "epoch": 7.0,
128
+ "eval_combined_score": 0.782515489635635,
129
+ "eval_loss": 0.9523068070411682,
130
+ "eval_pearson": 0.7821895116539266,
131
+ "eval_runtime": 0.4809,
132
+ "eval_samples_per_second": 3119.023,
133
+ "eval_spearmanr": 0.7828414676173435,
134
+ "eval_steps_per_second": 12.476,
135
+ "step": 161
136
+ },
137
+ {
138
+ "epoch": 8.0,
139
+ "grad_norm": 12.979435920715332,
140
+ "learning_rate": 4.2e-05,
141
+ "loss": 0.4506,
142
+ "step": 184
143
+ },
144
+ {
145
+ "epoch": 8.0,
146
+ "eval_combined_score": 0.7919260991828585,
147
+ "eval_loss": 0.9420682191848755,
148
+ "eval_pearson": 0.791584489351049,
149
+ "eval_runtime": 0.4809,
150
+ "eval_samples_per_second": 3118.926,
151
+ "eval_spearmanr": 0.7922677090146679,
152
+ "eval_steps_per_second": 12.476,
153
+ "step": 184
154
+ },
155
+ {
156
+ "epoch": 9.0,
157
+ "grad_norm": 18.948219299316406,
158
+ "learning_rate": 4.1e-05,
159
+ "loss": 0.4174,
160
+ "step": 207
161
+ },
162
+ {
163
+ "epoch": 9.0,
164
+ "eval_combined_score": 0.7906456677090563,
165
+ "eval_loss": 1.0924893617630005,
166
+ "eval_pearson": 0.7875795388915195,
167
+ "eval_runtime": 0.4739,
168
+ "eval_samples_per_second": 3165.183,
169
+ "eval_spearmanr": 0.7937117965265932,
170
+ "eval_steps_per_second": 12.661,
171
+ "step": 207
172
+ },
173
+ {
174
+ "epoch": 10.0,
175
+ "grad_norm": 5.4617791175842285,
176
+ "learning_rate": 4e-05,
177
+ "loss": 0.3229,
178
+ "step": 230
179
+ },
180
+ {
181
+ "epoch": 10.0,
182
+ "eval_combined_score": 0.7957522599726154,
183
+ "eval_loss": 0.8792051076889038,
184
+ "eval_pearson": 0.7969462774842241,
185
+ "eval_runtime": 0.4767,
186
+ "eval_samples_per_second": 3146.57,
187
+ "eval_spearmanr": 0.7945582424610067,
188
+ "eval_steps_per_second": 12.586,
189
+ "step": 230
190
+ },
191
+ {
192
+ "epoch": 11.0,
193
+ "grad_norm": 6.967871189117432,
194
+ "learning_rate": 3.9000000000000006e-05,
195
+ "loss": 0.3096,
196
+ "step": 253
197
+ },
198
+ {
199
+ "epoch": 11.0,
200
+ "eval_combined_score": 0.7885789869359312,
201
+ "eval_loss": 0.9665384292602539,
202
+ "eval_pearson": 0.7887171688488032,
203
+ "eval_runtime": 0.487,
204
+ "eval_samples_per_second": 3079.987,
205
+ "eval_spearmanr": 0.7884408050230591,
206
+ "eval_steps_per_second": 12.32,
207
+ "step": 253
208
+ },
209
+ {
210
+ "epoch": 12.0,
211
+ "grad_norm": 16.113603591918945,
212
+ "learning_rate": 3.8e-05,
213
+ "loss": 0.2918,
214
+ "step": 276
215
+ },
216
+ {
217
+ "epoch": 12.0,
218
+ "eval_combined_score": 0.8004322602860193,
219
+ "eval_loss": 0.8212088346481323,
220
+ "eval_pearson": 0.801677826027752,
221
+ "eval_runtime": 0.4993,
222
+ "eval_samples_per_second": 3003.968,
223
+ "eval_spearmanr": 0.7991866945442867,
224
+ "eval_steps_per_second": 12.016,
225
+ "step": 276
226
+ },
227
+ {
228
+ "epoch": 13.0,
229
+ "grad_norm": 4.451534271240234,
230
+ "learning_rate": 3.7e-05,
231
+ "loss": 0.2479,
232
+ "step": 299
233
+ },
234
+ {
235
+ "epoch": 13.0,
236
+ "eval_combined_score": 0.804614899302114,
237
+ "eval_loss": 0.8791233897209167,
238
+ "eval_pearson": 0.8055416707360773,
239
+ "eval_runtime": 0.4721,
240
+ "eval_samples_per_second": 3177.294,
241
+ "eval_spearmanr": 0.8036881278681508,
242
+ "eval_steps_per_second": 12.709,
243
+ "step": 299
244
+ },
245
+ {
246
+ "epoch": 14.0,
247
+ "grad_norm": 21.7775821685791,
248
+ "learning_rate": 3.6e-05,
249
+ "loss": 0.2409,
250
+ "step": 322
251
+ },
252
+ {
253
+ "epoch": 14.0,
254
+ "eval_combined_score": 0.794995597443876,
255
+ "eval_loss": 0.8217026591300964,
256
+ "eval_pearson": 0.7969938156234373,
257
+ "eval_runtime": 0.5148,
258
+ "eval_samples_per_second": 2913.928,
259
+ "eval_spearmanr": 0.7929973792643148,
260
+ "eval_steps_per_second": 11.656,
261
+ "step": 322
262
+ },
263
+ {
264
+ "epoch": 15.0,
265
+ "grad_norm": 3.07222056388855,
266
+ "learning_rate": 3.5e-05,
267
+ "loss": 0.2071,
268
+ "step": 345
269
+ },
270
+ {
271
+ "epoch": 15.0,
272
+ "eval_combined_score": 0.7984497715007259,
273
+ "eval_loss": 0.8929257988929749,
274
+ "eval_pearson": 0.7998156602749035,
275
+ "eval_runtime": 0.4892,
276
+ "eval_samples_per_second": 3066.293,
277
+ "eval_spearmanr": 0.7970838827265485,
278
+ "eval_steps_per_second": 12.265,
279
+ "step": 345
280
+ },
281
+ {
282
+ "epoch": 16.0,
283
+ "grad_norm": 3.4134459495544434,
284
+ "learning_rate": 3.4000000000000007e-05,
285
+ "loss": 0.1907,
286
+ "step": 368
287
+ },
288
+ {
289
+ "epoch": 16.0,
290
+ "eval_combined_score": 0.8007289150909647,
291
+ "eval_loss": 0.8188873529434204,
292
+ "eval_pearson": 0.8026761278764143,
293
+ "eval_runtime": 0.4814,
294
+ "eval_samples_per_second": 3116.2,
295
+ "eval_spearmanr": 0.798781702305515,
296
+ "eval_steps_per_second": 12.465,
297
+ "step": 368
298
+ },
299
+ {
300
+ "epoch": 17.0,
301
+ "grad_norm": 4.4020676612854,
302
+ "learning_rate": 3.3e-05,
303
+ "loss": 0.1825,
304
+ "step": 391
305
+ },
306
+ {
307
+ "epoch": 17.0,
308
+ "eval_combined_score": 0.79118899198846,
309
+ "eval_loss": 0.9392598271369934,
310
+ "eval_pearson": 0.7924365508807105,
311
+ "eval_runtime": 0.4766,
312
+ "eval_samples_per_second": 3147.292,
313
+ "eval_spearmanr": 0.7899414330962096,
314
+ "eval_steps_per_second": 12.589,
315
+ "step": 391
316
+ },
317
+ {
318
+ "epoch": 18.0,
319
+ "grad_norm": 5.791975498199463,
320
+ "learning_rate": 3.2000000000000005e-05,
321
+ "loss": 0.1634,
322
+ "step": 414
323
+ },
324
+ {
325
+ "epoch": 18.0,
326
+ "eval_combined_score": 0.8034539561261447,
327
+ "eval_loss": 0.8721528053283691,
328
+ "eval_pearson": 0.8045915853539458,
329
+ "eval_runtime": 0.4845,
330
+ "eval_samples_per_second": 3095.701,
331
+ "eval_spearmanr": 0.8023163268983434,
332
+ "eval_steps_per_second": 12.383,
333
+ "step": 414
334
+ },
335
+ {
336
+ "epoch": 19.0,
337
+ "grad_norm": 2.9870147705078125,
338
+ "learning_rate": 3.1e-05,
339
+ "loss": 0.1531,
340
+ "step": 437
341
+ },
342
+ {
343
+ "epoch": 19.0,
344
+ "eval_combined_score": 0.8030149651947462,
345
+ "eval_loss": 0.9629133343696594,
346
+ "eval_pearson": 0.8037084608596774,
347
+ "eval_runtime": 0.4776,
348
+ "eval_samples_per_second": 3140.565,
349
+ "eval_spearmanr": 0.8023214695298152,
350
+ "eval_steps_per_second": 12.562,
351
+ "step": 437
352
+ },
353
+ {
354
+ "epoch": 20.0,
355
+ "grad_norm": 3.91269850730896,
356
+ "learning_rate": 3e-05,
357
+ "loss": 0.153,
358
+ "step": 460
359
+ },
360
+ {
361
+ "epoch": 20.0,
362
+ "eval_combined_score": 0.7958772509210114,
363
+ "eval_loss": 1.0062919855117798,
364
+ "eval_pearson": 0.797094458559933,
365
+ "eval_runtime": 0.485,
366
+ "eval_samples_per_second": 3092.717,
367
+ "eval_spearmanr": 0.7946600432820897,
368
+ "eval_steps_per_second": 12.371,
369
+ "step": 460
370
+ },
371
+ {
372
+ "epoch": 21.0,
373
+ "grad_norm": 4.98586893081665,
374
+ "learning_rate": 2.9e-05,
375
+ "loss": 0.1409,
376
+ "step": 483
377
+ },
378
+ {
379
+ "epoch": 21.0,
380
+ "eval_combined_score": 0.7955185663928932,
381
+ "eval_loss": 0.9756875038146973,
382
+ "eval_pearson": 0.7968145893338887,
383
+ "eval_runtime": 0.478,
384
+ "eval_samples_per_second": 3138.129,
385
+ "eval_spearmanr": 0.7942225434518977,
386
+ "eval_steps_per_second": 12.553,
387
+ "step": 483
388
+ },
389
+ {
390
+ "epoch": 21.0,
391
+ "step": 483,
392
+ "total_flos": 3165851231294976.0,
393
+ "train_loss": 0.5645940970189823,
394
+ "train_runtime": 97.1506,
395
+ "train_samples_per_second": 2958.809,
396
+ "train_steps_per_second": 11.837
397
+ }
398
+ ],
399
+ "logging_steps": 1,
400
+ "max_steps": 1150,
401
+ "num_input_tokens_seen": 0,
402
+ "num_train_epochs": 50,
403
+ "save_steps": 500,
404
+ "stateful_callbacks": {
405
+ "EarlyStoppingCallback": {
406
+ "args": {
407
+ "early_stopping_patience": 5,
408
+ "early_stopping_threshold": 0.0
409
+ },
410
+ "attributes": {
411
+ "early_stopping_patience_counter": 5
412
+ }
413
+ },
414
+ "TrainerControl": {
415
+ "args": {
416
+ "should_epoch_stop": false,
417
+ "should_evaluate": false,
418
+ "should_log": false,
419
+ "should_save": true,
420
+ "should_training_stop": true
421
+ },
422
+ "attributes": {}
423
+ }
424
+ },
425
+ "total_flos": 3165851231294976.0,
426
+ "train_batch_size": 256,
427
+ "trial_name": null,
428
+ "trial_params": null
429
+ }