Md Mushfiqur Rahman commited on
Commit
ceb964c
1 Parent(s): 55b9859

Upload with huggingface_hub

Browse files
README.md ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - canine
5
+ - pretrained-on-english-language
6
+ ---
7
+
8
+ ### How to use
9
+
10
+ Here is how to use this model:
11
+
12
+ ```python
13
+ from transformers import CanineModel
14
+ model = CanineModel.from_pretrained('mushfiqur11/<repo name>')
15
+ ```
all_results.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 256.41,
3
+ "eval_accuracy": 0.899671052631579,
4
+ "eval_loss": 0.92512446641922,
5
+ "eval_runtime": 1.5393,
6
+ "eval_samples": 381,
7
+ "eval_samples_per_second": 247.514,
8
+ "eval_steps_per_second": 31.183,
9
+ "test_accuracy": 0.8918083462132921,
10
+ "test_loss": 0.9242635369300842,
11
+ "test_runtime": 1.6382,
12
+ "test_samples": 403,
13
+ "test_samples_per_second": 246.001,
14
+ "test_steps_per_second": 31.132,
15
+ "train_loss": 0.03826559160500765,
16
+ "train_runtime": 1428.3699,
17
+ "train_samples": 1227,
18
+ "train_samples_per_second": 336.047,
19
+ "train_steps_per_second": 10.501
20
+ }
config.json ADDED
@@ -0,0 +1,67 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/canine-s",
3
+ "architectures": [
4
+ "CanineForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 57344,
8
+ "downsampling_rate": 4,
9
+ "eos_token_id": 57345,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 768,
13
+ "id2label": {
14
+ "0": "ADJ",
15
+ "1": "ADP",
16
+ "2": "ADV",
17
+ "3": "AUX",
18
+ "4": "CCONJ",
19
+ "5": "DET",
20
+ "6": "INTJ",
21
+ "7": "NOUN",
22
+ "8": "NUM",
23
+ "9": "PART",
24
+ "10": "PRON",
25
+ "11": "PROPN",
26
+ "12": "PUNCT",
27
+ "13": "SCONJ",
28
+ "14": "SYM",
29
+ "15": "VERB",
30
+ "16": "X"
31
+ },
32
+ "initializer_range": 0.02,
33
+ "intermediate_size": 3072,
34
+ "label2id": {
35
+ "ADJ": 0,
36
+ "ADP": 1,
37
+ "ADV": 2,
38
+ "AUX": 3,
39
+ "CCONJ": 4,
40
+ "DET": 5,
41
+ "INTJ": 6,
42
+ "NOUN": 7,
43
+ "NUM": 8,
44
+ "PART": 9,
45
+ "PRON": 10,
46
+ "PROPN": 11,
47
+ "PUNCT": 12,
48
+ "SCONJ": 13,
49
+ "SYM": 14,
50
+ "VERB": 15,
51
+ "X": 16
52
+ },
53
+ "layer_norm_eps": 1e-12,
54
+ "local_transformer_stride": 128,
55
+ "max_position_embeddings": 16384,
56
+ "model_type": "canine",
57
+ "num_attention_heads": 12,
58
+ "num_hash_buckets": 16384,
59
+ "num_hash_functions": 8,
60
+ "num_hidden_layers": 12,
61
+ "pad_token_id": 0,
62
+ "torch_dtype": "float32",
63
+ "transformers_version": "4.17.0",
64
+ "type_vocab_size": 16,
65
+ "upsampling_kernel_size": 4,
66
+ "use_cache": true
67
+ }
eval_predictions.csv ADDED
The diff for this file is too large to render. See raw diff
 
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:003700cd673066901b796580165a04db7e01dc1033a9b28134b45c2b1995e8bf
3
+ size 528619377
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "eos_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "sep_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "pad_token": {"content": "\u0000", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "cls_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "mask_token": {"content": "", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true}}
test_predictions.csv ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "sep_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "cls_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "pad_token": {"content": "\u0000", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "mask_token": {"content": "", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": false, "model_max_length": 2048, "special_tokens_map_file": "/home/mrahma45/.cache/huggingface/transformers/00ffef2919c64d09945b6c06b1c03c4d18bf55f88013390d4443a1eb8c4f8a10.ab71f530366fe02e2834427e7b90198bfd0d573bc4279bfafdb2b95fe2b46dde", "name_or_path": "google/canine-s", "tokenizer_class": "CanineTokenizer"}
trainer_state.json ADDED
@@ -0,0 +1,805 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.899671052631579,
3
+ "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/canine/canine-base-finetuned-pos-ud-Coptic-Scriptorium/checkpoint-7500",
4
+ "epoch": 256.4102564102564,
5
+ "global_step": 10000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 2.56,
12
+ "learning_rate": 7.92e-05,
13
+ "loss": 1.66,
14
+ "step": 100
15
+ },
16
+ {
17
+ "epoch": 5.13,
18
+ "learning_rate": 7.946845637583894e-05,
19
+ "loss": 0.6182,
20
+ "step": 200
21
+ },
22
+ {
23
+ "epoch": 7.69,
24
+ "learning_rate": 7.893154362416109e-05,
25
+ "loss": 0.382,
26
+ "step": 300
27
+ },
28
+ {
29
+ "epoch": 10.26,
30
+ "learning_rate": 7.839463087248322e-05,
31
+ "loss": 0.2534,
32
+ "step": 400
33
+ },
34
+ {
35
+ "epoch": 12.82,
36
+ "learning_rate": 7.785771812080537e-05,
37
+ "loss": 0.1725,
38
+ "step": 500
39
+ },
40
+ {
41
+ "epoch": 12.82,
42
+ "eval_accuracy": 0.8799342105263158,
43
+ "eval_loss": 0.4891912639141083,
44
+ "eval_runtime": 1.5366,
45
+ "eval_samples_per_second": 247.947,
46
+ "eval_steps_per_second": 31.237,
47
+ "step": 500
48
+ },
49
+ {
50
+ "epoch": 15.38,
51
+ "learning_rate": 7.732080536912752e-05,
52
+ "loss": 0.1148,
53
+ "step": 600
54
+ },
55
+ {
56
+ "epoch": 17.95,
57
+ "learning_rate": 7.678389261744967e-05,
58
+ "loss": 0.0807,
59
+ "step": 700
60
+ },
61
+ {
62
+ "epoch": 20.51,
63
+ "learning_rate": 7.624697986577182e-05,
64
+ "loss": 0.0562,
65
+ "step": 800
66
+ },
67
+ {
68
+ "epoch": 23.08,
69
+ "learning_rate": 7.571006711409396e-05,
70
+ "loss": 0.0443,
71
+ "step": 900
72
+ },
73
+ {
74
+ "epoch": 25.64,
75
+ "learning_rate": 7.517315436241611e-05,
76
+ "loss": 0.0372,
77
+ "step": 1000
78
+ },
79
+ {
80
+ "epoch": 25.64,
81
+ "eval_accuracy": 0.8819444444444444,
82
+ "eval_loss": 0.6423949599266052,
83
+ "eval_runtime": 1.5561,
84
+ "eval_samples_per_second": 244.836,
85
+ "eval_steps_per_second": 30.845,
86
+ "step": 1000
87
+ },
88
+ {
89
+ "epoch": 28.21,
90
+ "learning_rate": 7.463624161073826e-05,
91
+ "loss": 0.0289,
92
+ "step": 1100
93
+ },
94
+ {
95
+ "epoch": 30.77,
96
+ "learning_rate": 7.409932885906041e-05,
97
+ "loss": 0.0236,
98
+ "step": 1200
99
+ },
100
+ {
101
+ "epoch": 33.33,
102
+ "learning_rate": 7.356241610738256e-05,
103
+ "loss": 0.0185,
104
+ "step": 1300
105
+ },
106
+ {
107
+ "epoch": 35.9,
108
+ "learning_rate": 7.30255033557047e-05,
109
+ "loss": 0.016,
110
+ "step": 1400
111
+ },
112
+ {
113
+ "epoch": 38.46,
114
+ "learning_rate": 7.248859060402685e-05,
115
+ "loss": 0.0161,
116
+ "step": 1500
117
+ },
118
+ {
119
+ "epoch": 38.46,
120
+ "eval_accuracy": 0.8866959064327485,
121
+ "eval_loss": 0.7260786890983582,
122
+ "eval_runtime": 1.5383,
123
+ "eval_samples_per_second": 247.68,
124
+ "eval_steps_per_second": 31.204,
125
+ "step": 1500
126
+ },
127
+ {
128
+ "epoch": 41.03,
129
+ "learning_rate": 7.1951677852349e-05,
130
+ "loss": 0.0142,
131
+ "step": 1600
132
+ },
133
+ {
134
+ "epoch": 43.59,
135
+ "learning_rate": 7.141476510067115e-05,
136
+ "loss": 0.0116,
137
+ "step": 1700
138
+ },
139
+ {
140
+ "epoch": 46.15,
141
+ "learning_rate": 7.08778523489933e-05,
142
+ "loss": 0.0117,
143
+ "step": 1800
144
+ },
145
+ {
146
+ "epoch": 48.72,
147
+ "learning_rate": 7.034093959731545e-05,
148
+ "loss": 0.0109,
149
+ "step": 1900
150
+ },
151
+ {
152
+ "epoch": 51.28,
153
+ "learning_rate": 6.98040268456376e-05,
154
+ "loss": 0.0096,
155
+ "step": 2000
156
+ },
157
+ {
158
+ "epoch": 51.28,
159
+ "eval_accuracy": 0.8880665204678363,
160
+ "eval_loss": 0.8069060444831848,
161
+ "eval_runtime": 1.5297,
162
+ "eval_samples_per_second": 249.069,
163
+ "eval_steps_per_second": 31.379,
164
+ "step": 2000
165
+ },
166
+ {
167
+ "epoch": 53.85,
168
+ "learning_rate": 6.926711409395974e-05,
169
+ "loss": 0.0086,
170
+ "step": 2100
171
+ },
172
+ {
173
+ "epoch": 56.41,
174
+ "learning_rate": 6.873020134228189e-05,
175
+ "loss": 0.0079,
176
+ "step": 2200
177
+ },
178
+ {
179
+ "epoch": 58.97,
180
+ "learning_rate": 6.819328859060404e-05,
181
+ "loss": 0.0069,
182
+ "step": 2300
183
+ },
184
+ {
185
+ "epoch": 61.54,
186
+ "learning_rate": 6.765637583892619e-05,
187
+ "loss": 0.0063,
188
+ "step": 2400
189
+ },
190
+ {
191
+ "epoch": 64.1,
192
+ "learning_rate": 6.711946308724834e-05,
193
+ "loss": 0.0075,
194
+ "step": 2500
195
+ },
196
+ {
197
+ "epoch": 64.1,
198
+ "eval_accuracy": 0.8915387426900585,
199
+ "eval_loss": 0.828658401966095,
200
+ "eval_runtime": 1.5268,
201
+ "eval_samples_per_second": 249.541,
202
+ "eval_steps_per_second": 31.438,
203
+ "step": 2500
204
+ },
205
+ {
206
+ "epoch": 66.67,
207
+ "learning_rate": 6.658255033557048e-05,
208
+ "loss": 0.0069,
209
+ "step": 2600
210
+ },
211
+ {
212
+ "epoch": 69.23,
213
+ "learning_rate": 6.604563758389262e-05,
214
+ "loss": 0.0071,
215
+ "step": 2700
216
+ },
217
+ {
218
+ "epoch": 71.79,
219
+ "learning_rate": 6.550872483221477e-05,
220
+ "loss": 0.0071,
221
+ "step": 2800
222
+ },
223
+ {
224
+ "epoch": 74.36,
225
+ "learning_rate": 6.497181208053692e-05,
226
+ "loss": 0.006,
227
+ "step": 2900
228
+ },
229
+ {
230
+ "epoch": 76.92,
231
+ "learning_rate": 6.443489932885906e-05,
232
+ "loss": 0.0058,
233
+ "step": 3000
234
+ },
235
+ {
236
+ "epoch": 76.92,
237
+ "eval_accuracy": 0.8901681286549707,
238
+ "eval_loss": 0.8421908020973206,
239
+ "eval_runtime": 1.5301,
240
+ "eval_samples_per_second": 249.0,
241
+ "eval_steps_per_second": 31.37,
242
+ "step": 3000
243
+ },
244
+ {
245
+ "epoch": 79.49,
246
+ "learning_rate": 6.389798657718121e-05,
247
+ "loss": 0.0064,
248
+ "step": 3100
249
+ },
250
+ {
251
+ "epoch": 82.05,
252
+ "learning_rate": 6.336107382550336e-05,
253
+ "loss": 0.0058,
254
+ "step": 3200
255
+ },
256
+ {
257
+ "epoch": 84.62,
258
+ "learning_rate": 6.282416107382551e-05,
259
+ "loss": 0.0061,
260
+ "step": 3300
261
+ },
262
+ {
263
+ "epoch": 87.18,
264
+ "learning_rate": 6.228724832214766e-05,
265
+ "loss": 0.0062,
266
+ "step": 3400
267
+ },
268
+ {
269
+ "epoch": 89.74,
270
+ "learning_rate": 6.17503355704698e-05,
271
+ "loss": 0.0052,
272
+ "step": 3500
273
+ },
274
+ {
275
+ "epoch": 89.74,
276
+ "eval_accuracy": 0.8907163742690059,
277
+ "eval_loss": 0.8915664553642273,
278
+ "eval_runtime": 1.5301,
279
+ "eval_samples_per_second": 248.996,
280
+ "eval_steps_per_second": 31.37,
281
+ "step": 3500
282
+ },
283
+ {
284
+ "epoch": 92.31,
285
+ "learning_rate": 6.121342281879195e-05,
286
+ "loss": 0.004,
287
+ "step": 3600
288
+ },
289
+ {
290
+ "epoch": 94.87,
291
+ "learning_rate": 6.06765100671141e-05,
292
+ "loss": 0.0055,
293
+ "step": 3700
294
+ },
295
+ {
296
+ "epoch": 97.44,
297
+ "learning_rate": 6.013959731543624e-05,
298
+ "loss": 0.0043,
299
+ "step": 3800
300
+ },
301
+ {
302
+ "epoch": 100.0,
303
+ "learning_rate": 5.960268456375839e-05,
304
+ "loss": 0.004,
305
+ "step": 3900
306
+ },
307
+ {
308
+ "epoch": 102.56,
309
+ "learning_rate": 5.906577181208054e-05,
310
+ "loss": 0.0036,
311
+ "step": 4000
312
+ },
313
+ {
314
+ "epoch": 102.56,
315
+ "eval_accuracy": 0.8918128654970761,
316
+ "eval_loss": 0.8925005793571472,
317
+ "eval_runtime": 1.5336,
318
+ "eval_samples_per_second": 248.438,
319
+ "eval_steps_per_second": 31.299,
320
+ "step": 4000
321
+ },
322
+ {
323
+ "epoch": 105.13,
324
+ "learning_rate": 5.852885906040269e-05,
325
+ "loss": 0.0044,
326
+ "step": 4100
327
+ },
328
+ {
329
+ "epoch": 107.69,
330
+ "learning_rate": 5.7991946308724836e-05,
331
+ "loss": 0.003,
332
+ "step": 4200
333
+ },
334
+ {
335
+ "epoch": 110.26,
336
+ "learning_rate": 5.7455033557046984e-05,
337
+ "loss": 0.0038,
338
+ "step": 4300
339
+ },
340
+ {
341
+ "epoch": 112.82,
342
+ "learning_rate": 5.691812080536913e-05,
343
+ "loss": 0.0034,
344
+ "step": 4400
345
+ },
346
+ {
347
+ "epoch": 115.38,
348
+ "learning_rate": 5.638120805369128e-05,
349
+ "loss": 0.0034,
350
+ "step": 4500
351
+ },
352
+ {
353
+ "epoch": 115.38,
354
+ "eval_accuracy": 0.8918128654970761,
355
+ "eval_loss": 0.9117233753204346,
356
+ "eval_runtime": 1.5348,
357
+ "eval_samples_per_second": 248.239,
358
+ "eval_steps_per_second": 31.274,
359
+ "step": 4500
360
+ },
361
+ {
362
+ "epoch": 117.95,
363
+ "learning_rate": 5.584429530201343e-05,
364
+ "loss": 0.0039,
365
+ "step": 4600
366
+ },
367
+ {
368
+ "epoch": 120.51,
369
+ "learning_rate": 5.530738255033558e-05,
370
+ "loss": 0.0037,
371
+ "step": 4700
372
+ },
373
+ {
374
+ "epoch": 123.08,
375
+ "learning_rate": 5.4770469798657725e-05,
376
+ "loss": 0.0031,
377
+ "step": 4800
378
+ },
379
+ {
380
+ "epoch": 125.64,
381
+ "learning_rate": 5.4233557046979874e-05,
382
+ "loss": 0.0038,
383
+ "step": 4900
384
+ },
385
+ {
386
+ "epoch": 128.21,
387
+ "learning_rate": 5.3696644295302015e-05,
388
+ "loss": 0.004,
389
+ "step": 5000
390
+ },
391
+ {
392
+ "epoch": 128.21,
393
+ "eval_accuracy": 0.8946454678362573,
394
+ "eval_loss": 0.8941249847412109,
395
+ "eval_runtime": 1.5416,
396
+ "eval_samples_per_second": 247.142,
397
+ "eval_steps_per_second": 31.136,
398
+ "step": 5000
399
+ },
400
+ {
401
+ "epoch": 130.77,
402
+ "learning_rate": 5.315973154362416e-05,
403
+ "loss": 0.0025,
404
+ "step": 5100
405
+ },
406
+ {
407
+ "epoch": 133.33,
408
+ "learning_rate": 5.262281879194631e-05,
409
+ "loss": 0.0025,
410
+ "step": 5200
411
+ },
412
+ {
413
+ "epoch": 135.9,
414
+ "learning_rate": 5.208590604026846e-05,
415
+ "loss": 0.0035,
416
+ "step": 5300
417
+ },
418
+ {
419
+ "epoch": 138.46,
420
+ "learning_rate": 5.154899328859061e-05,
421
+ "loss": 0.0028,
422
+ "step": 5400
423
+ },
424
+ {
425
+ "epoch": 141.03,
426
+ "learning_rate": 5.1012080536912756e-05,
427
+ "loss": 0.0033,
428
+ "step": 5500
429
+ },
430
+ {
431
+ "epoch": 141.03,
432
+ "eval_accuracy": 0.8901681286549707,
433
+ "eval_loss": 0.9474687576293945,
434
+ "eval_runtime": 1.5483,
435
+ "eval_samples_per_second": 246.083,
436
+ "eval_steps_per_second": 31.003,
437
+ "step": 5500
438
+ },
439
+ {
440
+ "epoch": 143.59,
441
+ "learning_rate": 5.0475167785234905e-05,
442
+ "loss": 0.0026,
443
+ "step": 5600
444
+ },
445
+ {
446
+ "epoch": 146.15,
447
+ "learning_rate": 4.993825503355705e-05,
448
+ "loss": 0.0021,
449
+ "step": 5700
450
+ },
451
+ {
452
+ "epoch": 148.72,
453
+ "learning_rate": 4.94013422818792e-05,
454
+ "loss": 0.0025,
455
+ "step": 5800
456
+ },
457
+ {
458
+ "epoch": 151.28,
459
+ "learning_rate": 4.886442953020135e-05,
460
+ "loss": 0.002,
461
+ "step": 5900
462
+ },
463
+ {
464
+ "epoch": 153.85,
465
+ "learning_rate": 4.83275167785235e-05,
466
+ "loss": 0.0028,
467
+ "step": 6000
468
+ },
469
+ {
470
+ "epoch": 153.85,
471
+ "eval_accuracy": 0.8967470760233918,
472
+ "eval_loss": 0.9054576754570007,
473
+ "eval_runtime": 1.5492,
474
+ "eval_samples_per_second": 245.926,
475
+ "eval_steps_per_second": 30.983,
476
+ "step": 6000
477
+ },
478
+ {
479
+ "epoch": 156.41,
480
+ "learning_rate": 4.779060402684564e-05,
481
+ "loss": 0.0026,
482
+ "step": 6100
483
+ },
484
+ {
485
+ "epoch": 158.97,
486
+ "learning_rate": 4.725369127516779e-05,
487
+ "loss": 0.0022,
488
+ "step": 6200
489
+ },
490
+ {
491
+ "epoch": 161.54,
492
+ "learning_rate": 4.6716778523489936e-05,
493
+ "loss": 0.0026,
494
+ "step": 6300
495
+ },
496
+ {
497
+ "epoch": 164.1,
498
+ "learning_rate": 4.6179865771812084e-05,
499
+ "loss": 0.0025,
500
+ "step": 6400
501
+ },
502
+ {
503
+ "epoch": 166.67,
504
+ "learning_rate": 4.564295302013423e-05,
505
+ "loss": 0.0014,
506
+ "step": 6500
507
+ },
508
+ {
509
+ "epoch": 166.67,
510
+ "eval_accuracy": 0.8951937134502924,
511
+ "eval_loss": 0.9585675597190857,
512
+ "eval_runtime": 1.544,
513
+ "eval_samples_per_second": 246.764,
514
+ "eval_steps_per_second": 31.088,
515
+ "step": 6500
516
+ },
517
+ {
518
+ "epoch": 169.23,
519
+ "learning_rate": 4.510604026845638e-05,
520
+ "loss": 0.0027,
521
+ "step": 6600
522
+ },
523
+ {
524
+ "epoch": 171.79,
525
+ "learning_rate": 4.456912751677853e-05,
526
+ "loss": 0.0016,
527
+ "step": 6700
528
+ },
529
+ {
530
+ "epoch": 174.36,
531
+ "learning_rate": 4.403221476510068e-05,
532
+ "loss": 0.0024,
533
+ "step": 6800
534
+ },
535
+ {
536
+ "epoch": 176.92,
537
+ "learning_rate": 4.3495302013422825e-05,
538
+ "loss": 0.0017,
539
+ "step": 6900
540
+ },
541
+ {
542
+ "epoch": 179.49,
543
+ "learning_rate": 4.295838926174497e-05,
544
+ "loss": 0.0018,
545
+ "step": 7000
546
+ },
547
+ {
548
+ "epoch": 179.49,
549
+ "eval_accuracy": 0.89546783625731,
550
+ "eval_loss": 0.9691464900970459,
551
+ "eval_runtime": 1.518,
552
+ "eval_samples_per_second": 250.996,
553
+ "eval_steps_per_second": 31.622,
554
+ "step": 7000
555
+ },
556
+ {
557
+ "epoch": 182.05,
558
+ "learning_rate": 4.242147651006712e-05,
559
+ "loss": 0.0023,
560
+ "step": 7100
561
+ },
562
+ {
563
+ "epoch": 184.62,
564
+ "learning_rate": 4.188456375838927e-05,
565
+ "loss": 0.0027,
566
+ "step": 7200
567
+ },
568
+ {
569
+ "epoch": 187.18,
570
+ "learning_rate": 4.134765100671141e-05,
571
+ "loss": 0.0021,
572
+ "step": 7300
573
+ },
574
+ {
575
+ "epoch": 189.74,
576
+ "learning_rate": 4.081073825503356e-05,
577
+ "loss": 0.002,
578
+ "step": 7400
579
+ },
580
+ {
581
+ "epoch": 192.31,
582
+ "learning_rate": 4.027382550335571e-05,
583
+ "loss": 0.0017,
584
+ "step": 7500
585
+ },
586
+ {
587
+ "epoch": 192.31,
588
+ "eval_accuracy": 0.899671052631579,
589
+ "eval_loss": 0.92512446641922,
590
+ "eval_runtime": 1.5211,
591
+ "eval_samples_per_second": 250.471,
592
+ "eval_steps_per_second": 31.555,
593
+ "step": 7500
594
+ },
595
+ {
596
+ "epoch": 194.87,
597
+ "learning_rate": 3.9736912751677856e-05,
598
+ "loss": 0.0017,
599
+ "step": 7600
600
+ },
601
+ {
602
+ "epoch": 197.44,
603
+ "learning_rate": 3.9200000000000004e-05,
604
+ "loss": 0.0019,
605
+ "step": 7700
606
+ },
607
+ {
608
+ "epoch": 200.0,
609
+ "learning_rate": 3.866308724832215e-05,
610
+ "loss": 0.0022,
611
+ "step": 7800
612
+ },
613
+ {
614
+ "epoch": 202.56,
615
+ "learning_rate": 3.81261744966443e-05,
616
+ "loss": 0.0012,
617
+ "step": 7900
618
+ },
619
+ {
620
+ "epoch": 205.13,
621
+ "learning_rate": 3.758926174496645e-05,
622
+ "loss": 0.0016,
623
+ "step": 8000
624
+ },
625
+ {
626
+ "epoch": 205.13,
627
+ "eval_accuracy": 0.8972953216374269,
628
+ "eval_loss": 0.9501896500587463,
629
+ "eval_runtime": 1.5504,
630
+ "eval_samples_per_second": 245.737,
631
+ "eval_steps_per_second": 30.959,
632
+ "step": 8000
633
+ },
634
+ {
635
+ "epoch": 207.69,
636
+ "learning_rate": 3.705234899328859e-05,
637
+ "loss": 0.0013,
638
+ "step": 8100
639
+ },
640
+ {
641
+ "epoch": 210.26,
642
+ "learning_rate": 3.651543624161074e-05,
643
+ "loss": 0.0011,
644
+ "step": 8200
645
+ },
646
+ {
647
+ "epoch": 212.82,
648
+ "learning_rate": 3.597852348993289e-05,
649
+ "loss": 0.0015,
650
+ "step": 8300
651
+ },
652
+ {
653
+ "epoch": 215.38,
654
+ "learning_rate": 3.5441610738255035e-05,
655
+ "loss": 0.0009,
656
+ "step": 8400
657
+ },
658
+ {
659
+ "epoch": 217.95,
660
+ "learning_rate": 3.490469798657718e-05,
661
+ "loss": 0.0015,
662
+ "step": 8500
663
+ },
664
+ {
665
+ "epoch": 217.95,
666
+ "eval_accuracy": 0.8966557017543859,
667
+ "eval_loss": 0.9619430303573608,
668
+ "eval_runtime": 1.5407,
669
+ "eval_samples_per_second": 247.295,
670
+ "eval_steps_per_second": 31.155,
671
+ "step": 8500
672
+ },
673
+ {
674
+ "epoch": 220.51,
675
+ "learning_rate": 3.436778523489933e-05,
676
+ "loss": 0.0011,
677
+ "step": 8600
678
+ },
679
+ {
680
+ "epoch": 223.08,
681
+ "learning_rate": 3.383087248322148e-05,
682
+ "loss": 0.002,
683
+ "step": 8700
684
+ },
685
+ {
686
+ "epoch": 225.64,
687
+ "learning_rate": 3.329395973154363e-05,
688
+ "loss": 0.0012,
689
+ "step": 8800
690
+ },
691
+ {
692
+ "epoch": 228.21,
693
+ "learning_rate": 3.2757046979865776e-05,
694
+ "loss": 0.0011,
695
+ "step": 8900
696
+ },
697
+ {
698
+ "epoch": 230.77,
699
+ "learning_rate": 3.2220134228187925e-05,
700
+ "loss": 0.001,
701
+ "step": 9000
702
+ },
703
+ {
704
+ "epoch": 230.77,
705
+ "eval_accuracy": 0.899671052631579,
706
+ "eval_loss": 0.9450499415397644,
707
+ "eval_runtime": 1.5395,
708
+ "eval_samples_per_second": 247.489,
709
+ "eval_steps_per_second": 31.18,
710
+ "step": 9000
711
+ },
712
+ {
713
+ "epoch": 233.33,
714
+ "learning_rate": 3.168322147651007e-05,
715
+ "loss": 0.0009,
716
+ "step": 9100
717
+ },
718
+ {
719
+ "epoch": 235.9,
720
+ "learning_rate": 3.114630872483222e-05,
721
+ "loss": 0.0009,
722
+ "step": 9200
723
+ },
724
+ {
725
+ "epoch": 238.46,
726
+ "learning_rate": 3.060939597315436e-05,
727
+ "loss": 0.0011,
728
+ "step": 9300
729
+ },
730
+ {
731
+ "epoch": 241.03,
732
+ "learning_rate": 3.007248322147651e-05,
733
+ "loss": 0.0006,
734
+ "step": 9400
735
+ },
736
+ {
737
+ "epoch": 243.59,
738
+ "learning_rate": 2.953557046979866e-05,
739
+ "loss": 0.0011,
740
+ "step": 9500
741
+ },
742
+ {
743
+ "epoch": 243.59,
744
+ "eval_accuracy": 0.8972953216374269,
745
+ "eval_loss": 0.9794738292694092,
746
+ "eval_runtime": 1.5217,
747
+ "eval_samples_per_second": 250.373,
748
+ "eval_steps_per_second": 31.543,
749
+ "step": 9500
750
+ },
751
+ {
752
+ "epoch": 246.15,
753
+ "learning_rate": 2.8998657718120807e-05,
754
+ "loss": 0.0009,
755
+ "step": 9600
756
+ },
757
+ {
758
+ "epoch": 248.72,
759
+ "learning_rate": 2.8461744966442955e-05,
760
+ "loss": 0.0009,
761
+ "step": 9700
762
+ },
763
+ {
764
+ "epoch": 251.28,
765
+ "learning_rate": 2.79248322147651e-05,
766
+ "loss": 0.0008,
767
+ "step": 9800
768
+ },
769
+ {
770
+ "epoch": 253.85,
771
+ "learning_rate": 2.738791946308725e-05,
772
+ "loss": 0.0004,
773
+ "step": 9900
774
+ },
775
+ {
776
+ "epoch": 256.41,
777
+ "learning_rate": 2.6851006711409397e-05,
778
+ "loss": 0.0011,
779
+ "step": 10000
780
+ },
781
+ {
782
+ "epoch": 256.41,
783
+ "eval_accuracy": 0.8970211988304093,
784
+ "eval_loss": 1.0004897117614746,
785
+ "eval_runtime": 1.5228,
786
+ "eval_samples_per_second": 250.203,
787
+ "eval_steps_per_second": 31.522,
788
+ "step": 10000
789
+ },
790
+ {
791
+ "epoch": 256.41,
792
+ "step": 10000,
793
+ "total_flos": 5.166941021601792e+16,
794
+ "train_loss": 0.03826559160500765,
795
+ "train_runtime": 1428.3699,
796
+ "train_samples_per_second": 336.047,
797
+ "train_steps_per_second": 10.501
798
+ }
799
+ ],
800
+ "max_steps": 15000,
801
+ "num_train_epochs": 385,
802
+ "total_flos": 5.166941021601792e+16,
803
+ "trial_name": null,
804
+ "trial_params": null
805
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:21e9647ed11e3445eb8170f54e6e3a419997b56d2fd5cc4fe75ed29b51e4a176
3
+ size 3387