xyzheng1995 commited on
Commit
179cc5d
1 Parent(s): b403e17

End of training

Browse files
README.md CHANGED
@@ -1,8 +1,9 @@
1
  ---
 
2
  tags:
3
  - generated_from_trainer
4
  datasets:
5
- - coco_dataset_script
6
  model-index:
7
  - name: clip-roberta-finetuned
8
  results: []
@@ -13,7 +14,9 @@ should probably proofread and complete it, then remove this comment. -->
13
 
14
  # clip-roberta-finetuned
15
 
16
- This model was trained from scratch on the coco_dataset_script dataset.
 
 
17
 
18
  ## Model description
19
 
 
1
  ---
2
+ base_model: ./clip-roberta
3
  tags:
4
  - generated_from_trainer
5
  datasets:
6
+ - ydshieh/coco_dataset_script
7
  model-index:
8
  - name: clip-roberta-finetuned
9
  results: []
 
14
 
15
  # clip-roberta-finetuned
16
 
17
+ This model is a fine-tuned version of [./clip-roberta](https://huggingface.co/./clip-roberta) on the ydshieh/coco_dataset_script 2017 dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 1.6958
20
 
21
  ## Model description
22
 
all_results.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_loss": 1.6957839727401733,
4
+ "eval_runtime": 166.0767,
5
+ "eval_samples_per_second": 150.617,
6
+ "eval_steps_per_second": 2.354,
7
+ "train_loss": 0.38465954929881363,
8
+ "train_runtime": 19859.5179,
9
+ "train_samples_per_second": 89.391,
10
+ "train_steps_per_second": 1.397
11
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_loss": 1.6957839727401733,
4
+ "eval_runtime": 166.0767,
5
+ "eval_samples_per_second": 150.617,
6
+ "eval_steps_per_second": 2.354
7
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
preprocessor_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "processor_class": "VisionTextDualEncoderProcessor",
23
+ "resample": 3,
24
+ "rescale_factor": 0.00392156862745098,
25
+ "size": {
26
+ "shortest_edge": 224
27
+ }
28
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "bos_token": "<s>",
4
+ "clean_up_tokenization_spaces": true,
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "errors": "replace",
8
+ "mask_token": "<mask>",
9
+ "model_max_length": 512,
10
+ "pad_token": "<pad>",
11
+ "processor_class": "VisionTextDualEncoderProcessor",
12
+ "sep_token": "</s>",
13
+ "tokenizer_class": "RobertaTokenizer",
14
+ "trim_offsets": true,
15
+ "unk_token": "<unk>"
16
+ }
train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 0.38465954929881363,
4
+ "train_runtime": 19859.5179,
5
+ "train_samples_per_second": 89.391,
6
+ "train_steps_per_second": 1.397
7
+ }
trainer_state.json ADDED
@@ -0,0 +1,358 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 27741,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.05,
13
+ "learning_rate": 4.909880682022999e-05,
14
+ "loss": 1.6412,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.11,
19
+ "learning_rate": 4.819761364045997e-05,
20
+ "loss": 0.9506,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.16,
25
+ "learning_rate": 4.7296420460689956e-05,
26
+ "loss": 0.8486,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 0.22,
31
+ "learning_rate": 4.639522728091994e-05,
32
+ "loss": 0.7927,
33
+ "step": 2000
34
+ },
35
+ {
36
+ "epoch": 0.27,
37
+ "learning_rate": 4.5494034101149925e-05,
38
+ "loss": 0.7523,
39
+ "step": 2500
40
+ },
41
+ {
42
+ "epoch": 0.32,
43
+ "learning_rate": 4.459284092137991e-05,
44
+ "loss": 0.7119,
45
+ "step": 3000
46
+ },
47
+ {
48
+ "epoch": 0.38,
49
+ "learning_rate": 4.3691647741609894e-05,
50
+ "loss": 0.6613,
51
+ "step": 3500
52
+ },
53
+ {
54
+ "epoch": 0.43,
55
+ "learning_rate": 4.279045456183988e-05,
56
+ "loss": 0.6448,
57
+ "step": 4000
58
+ },
59
+ {
60
+ "epoch": 0.49,
61
+ "learning_rate": 4.188926138206986e-05,
62
+ "loss": 0.6113,
63
+ "step": 4500
64
+ },
65
+ {
66
+ "epoch": 0.54,
67
+ "learning_rate": 4.098806820229985e-05,
68
+ "loss": 0.5901,
69
+ "step": 5000
70
+ },
71
+ {
72
+ "epoch": 0.59,
73
+ "learning_rate": 4.008687502252983e-05,
74
+ "loss": 0.5656,
75
+ "step": 5500
76
+ },
77
+ {
78
+ "epoch": 0.65,
79
+ "learning_rate": 3.9185681842759816e-05,
80
+ "loss": 0.541,
81
+ "step": 6000
82
+ },
83
+ {
84
+ "epoch": 0.7,
85
+ "learning_rate": 3.82844886629898e-05,
86
+ "loss": 0.527,
87
+ "step": 6500
88
+ },
89
+ {
90
+ "epoch": 0.76,
91
+ "learning_rate": 3.7383295483219785e-05,
92
+ "loss": 0.5115,
93
+ "step": 7000
94
+ },
95
+ {
96
+ "epoch": 0.81,
97
+ "learning_rate": 3.648210230344977e-05,
98
+ "loss": 0.4834,
99
+ "step": 7500
100
+ },
101
+ {
102
+ "epoch": 0.87,
103
+ "learning_rate": 3.558090912367975e-05,
104
+ "loss": 0.4699,
105
+ "step": 8000
106
+ },
107
+ {
108
+ "epoch": 0.92,
109
+ "learning_rate": 3.467971594390974e-05,
110
+ "loss": 0.4512,
111
+ "step": 8500
112
+ },
113
+ {
114
+ "epoch": 0.97,
115
+ "learning_rate": 3.3778522764139723e-05,
116
+ "loss": 0.4435,
117
+ "step": 9000
118
+ },
119
+ {
120
+ "epoch": 1.03,
121
+ "learning_rate": 3.287732958436971e-05,
122
+ "loss": 0.4027,
123
+ "step": 9500
124
+ },
125
+ {
126
+ "epoch": 1.08,
127
+ "learning_rate": 3.197613640459969e-05,
128
+ "loss": 0.366,
129
+ "step": 10000
130
+ },
131
+ {
132
+ "epoch": 1.14,
133
+ "learning_rate": 3.107494322482968e-05,
134
+ "loss": 0.36,
135
+ "step": 10500
136
+ },
137
+ {
138
+ "epoch": 1.19,
139
+ "learning_rate": 3.017375004505966e-05,
140
+ "loss": 0.3478,
141
+ "step": 11000
142
+ },
143
+ {
144
+ "epoch": 1.24,
145
+ "learning_rate": 2.9272556865289646e-05,
146
+ "loss": 0.3397,
147
+ "step": 11500
148
+ },
149
+ {
150
+ "epoch": 1.3,
151
+ "learning_rate": 2.837136368551963e-05,
152
+ "loss": 0.3336,
153
+ "step": 12000
154
+ },
155
+ {
156
+ "epoch": 1.35,
157
+ "learning_rate": 2.7470170505749615e-05,
158
+ "loss": 0.3313,
159
+ "step": 12500
160
+ },
161
+ {
162
+ "epoch": 1.41,
163
+ "learning_rate": 2.65689773259796e-05,
164
+ "loss": 0.3166,
165
+ "step": 13000
166
+ },
167
+ {
168
+ "epoch": 1.46,
169
+ "learning_rate": 2.5667784146209584e-05,
170
+ "loss": 0.3219,
171
+ "step": 13500
172
+ },
173
+ {
174
+ "epoch": 1.51,
175
+ "learning_rate": 2.476659096643957e-05,
176
+ "loss": 0.3166,
177
+ "step": 14000
178
+ },
179
+ {
180
+ "epoch": 1.57,
181
+ "learning_rate": 2.3865397786669553e-05,
182
+ "loss": 0.2962,
183
+ "step": 14500
184
+ },
185
+ {
186
+ "epoch": 1.62,
187
+ "learning_rate": 2.2964204606899537e-05,
188
+ "loss": 0.2987,
189
+ "step": 15000
190
+ },
191
+ {
192
+ "epoch": 1.68,
193
+ "learning_rate": 2.2063011427129522e-05,
194
+ "loss": 0.2902,
195
+ "step": 15500
196
+ },
197
+ {
198
+ "epoch": 1.73,
199
+ "learning_rate": 2.1161818247359506e-05,
200
+ "loss": 0.2802,
201
+ "step": 16000
202
+ },
203
+ {
204
+ "epoch": 1.78,
205
+ "learning_rate": 2.026062506758949e-05,
206
+ "loss": 0.2785,
207
+ "step": 16500
208
+ },
209
+ {
210
+ "epoch": 1.84,
211
+ "learning_rate": 1.9359431887819472e-05,
212
+ "loss": 0.2664,
213
+ "step": 17000
214
+ },
215
+ {
216
+ "epoch": 1.89,
217
+ "learning_rate": 1.8458238708049457e-05,
218
+ "loss": 0.2704,
219
+ "step": 17500
220
+ },
221
+ {
222
+ "epoch": 1.95,
223
+ "learning_rate": 1.755704552827944e-05,
224
+ "loss": 0.2624,
225
+ "step": 18000
226
+ },
227
+ {
228
+ "epoch": 2.0,
229
+ "learning_rate": 1.6655852348509426e-05,
230
+ "loss": 0.2504,
231
+ "step": 18500
232
+ },
233
+ {
234
+ "epoch": 2.05,
235
+ "learning_rate": 1.575465916873941e-05,
236
+ "loss": 0.2024,
237
+ "step": 19000
238
+ },
239
+ {
240
+ "epoch": 2.11,
241
+ "learning_rate": 1.4853465988969395e-05,
242
+ "loss": 0.198,
243
+ "step": 19500
244
+ },
245
+ {
246
+ "epoch": 2.16,
247
+ "learning_rate": 1.3952272809199379e-05,
248
+ "loss": 0.1896,
249
+ "step": 20000
250
+ },
251
+ {
252
+ "epoch": 2.22,
253
+ "learning_rate": 1.3051079629429367e-05,
254
+ "loss": 0.1922,
255
+ "step": 20500
256
+ },
257
+ {
258
+ "epoch": 2.27,
259
+ "learning_rate": 1.214988644965935e-05,
260
+ "loss": 0.187,
261
+ "step": 21000
262
+ },
263
+ {
264
+ "epoch": 2.33,
265
+ "learning_rate": 1.1248693269889334e-05,
266
+ "loss": 0.1809,
267
+ "step": 21500
268
+ },
269
+ {
270
+ "epoch": 2.38,
271
+ "learning_rate": 1.0347500090119317e-05,
272
+ "loss": 0.1856,
273
+ "step": 22000
274
+ },
275
+ {
276
+ "epoch": 2.43,
277
+ "learning_rate": 9.446306910349303e-06,
278
+ "loss": 0.1783,
279
+ "step": 22500
280
+ },
281
+ {
282
+ "epoch": 2.49,
283
+ "learning_rate": 8.545113730579288e-06,
284
+ "loss": 0.1743,
285
+ "step": 23000
286
+ },
287
+ {
288
+ "epoch": 2.54,
289
+ "learning_rate": 7.643920550809272e-06,
290
+ "loss": 0.1725,
291
+ "step": 23500
292
+ },
293
+ {
294
+ "epoch": 2.6,
295
+ "learning_rate": 6.742727371039257e-06,
296
+ "loss": 0.171,
297
+ "step": 24000
298
+ },
299
+ {
300
+ "epoch": 2.65,
301
+ "learning_rate": 5.84153419126924e-06,
302
+ "loss": 0.1595,
303
+ "step": 24500
304
+ },
305
+ {
306
+ "epoch": 2.7,
307
+ "learning_rate": 4.940341011499225e-06,
308
+ "loss": 0.1634,
309
+ "step": 25000
310
+ },
311
+ {
312
+ "epoch": 2.76,
313
+ "learning_rate": 4.03914783172921e-06,
314
+ "loss": 0.1618,
315
+ "step": 25500
316
+ },
317
+ {
318
+ "epoch": 2.81,
319
+ "learning_rate": 3.1379546519591943e-06,
320
+ "loss": 0.1607,
321
+ "step": 26000
322
+ },
323
+ {
324
+ "epoch": 2.87,
325
+ "learning_rate": 2.2367614721891784e-06,
326
+ "loss": 0.1557,
327
+ "step": 26500
328
+ },
329
+ {
330
+ "epoch": 2.92,
331
+ "learning_rate": 1.335568292419163e-06,
332
+ "loss": 0.1569,
333
+ "step": 27000
334
+ },
335
+ {
336
+ "epoch": 2.97,
337
+ "learning_rate": 4.3437511264914753e-07,
338
+ "loss": 0.1539,
339
+ "step": 27500
340
+ },
341
+ {
342
+ "epoch": 3.0,
343
+ "step": 27741,
344
+ "total_flos": 2.3702775297552e+17,
345
+ "train_loss": 0.38465954929881363,
346
+ "train_runtime": 19859.5179,
347
+ "train_samples_per_second": 89.391,
348
+ "train_steps_per_second": 1.397
349
+ }
350
+ ],
351
+ "logging_steps": 500,
352
+ "max_steps": 27741,
353
+ "num_train_epochs": 3,
354
+ "save_steps": 500,
355
+ "total_flos": 2.3702775297552e+17,
356
+ "trial_name": null,
357
+ "trial_params": null
358
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff