infinite000 commited on
Commit
fca301b
·
verified ·
1 Parent(s): b13c79c

Delete trainer_state.json

Browse files
Files changed (1) hide show
  1. trainer_state.json +0 -354
trainer_state.json DELETED
@@ -1,354 +0,0 @@
1
- {
2
- "best_global_step": 1035,
3
- "best_metric": 0.9635036496350365,
4
- "best_model_checkpoint": "/home/user/natix-mining-project/models/dinov2_roadwork/checkpoint-1035",
5
- "epoch": 15.0,
6
- "eval_steps": 500,
7
- "global_step": 1035,
8
- "is_hyper_param_search": false,
9
- "is_local_process_zero": true,
10
- "is_world_process_zero": true,
11
- "log_history": [
12
- {
13
- "epoch": 0.7299270072992701,
14
- "grad_norm": 16.808786392211914,
15
- "learning_rate": 3.3333333333333335e-07,
16
- "loss": 0.5872037124633789,
17
- "step": 50
18
- },
19
- {
20
- "epoch": 1.0,
21
- "eval_accuracy": 0.8219616204690832,
22
- "eval_f1": 0.8982327848872639,
23
- "eval_loss": 0.4097755253314972,
24
- "eval_precision": 0.8318284424379232,
25
- "eval_recall": 0.976158940397351,
26
- "eval_runtime": 87.0676,
27
- "eval_samples_per_second": 10.773,
28
- "eval_steps_per_second": 0.345,
29
- "step": 69
30
- },
31
- {
32
- "epoch": 1.4525547445255476,
33
- "grad_norm": 26.81678581237793,
34
- "learning_rate": 6.884057971014494e-07,
35
- "loss": 0.420146598815918,
36
- "step": 100
37
- },
38
- {
39
- "epoch": 2.0,
40
- "eval_accuracy": 0.908315565031983,
41
- "eval_f1": 0.944372574385511,
42
- "eval_loss": 0.23657475411891937,
43
- "eval_precision": 0.922882427307206,
44
- "eval_recall": 0.9668874172185431,
45
- "eval_runtime": 89.7516,
46
- "eval_samples_per_second": 10.451,
47
- "eval_steps_per_second": 0.334,
48
- "step": 138
49
- },
50
- {
51
- "epoch": 2.1751824817518246,
52
- "grad_norm": 17.34227180480957,
53
- "learning_rate": 1.0434782608695653e-06,
54
- "loss": 0.27175182342529297,
55
- "step": 150
56
- },
57
- {
58
- "epoch": 2.905109489051095,
59
- "grad_norm": 19.6271915435791,
60
- "learning_rate": 1.4057971014492755e-06,
61
- "loss": 0.23565080642700195,
62
- "step": 200
63
- },
64
- {
65
- "epoch": 3.0,
66
- "eval_accuracy": 0.9168443496801706,
67
- "eval_f1": 0.9483443708609272,
68
- "eval_loss": 0.20701788365840912,
69
- "eval_precision": 0.9483443708609272,
70
- "eval_recall": 0.9483443708609272,
71
- "eval_runtime": 83.376,
72
- "eval_samples_per_second": 11.25,
73
- "eval_steps_per_second": 0.36,
74
- "step": 207
75
- },
76
- {
77
- "epoch": 3.627737226277372,
78
- "grad_norm": 19.950403213500977,
79
- "learning_rate": 1.7681159420289855e-06,
80
- "loss": 0.2077177619934082,
81
- "step": 250
82
- },
83
- {
84
- "epoch": 4.0,
85
- "eval_accuracy": 0.9275053304904051,
86
- "eval_f1": 0.9554390563564875,
87
- "eval_loss": 0.1865338534116745,
88
- "eval_precision": 0.9455252918287937,
89
- "eval_recall": 0.9655629139072848,
90
- "eval_runtime": 95.1442,
91
- "eval_samples_per_second": 9.859,
92
- "eval_steps_per_second": 0.315,
93
- "step": 276
94
- },
95
- {
96
- "epoch": 4.350364963503649,
97
- "grad_norm": 14.074359893798828,
98
- "learning_rate": 2.130434782608696e-06,
99
- "loss": 0.1941886329650879,
100
- "step": 300
101
- },
102
- {
103
- "epoch": 5.0,
104
- "eval_accuracy": 0.9392324093816631,
105
- "eval_f1": 0.9623762376237623,
106
- "eval_loss": 0.1691393107175827,
107
- "eval_precision": 0.9592105263157895,
108
- "eval_recall": 0.9655629139072848,
109
- "eval_runtime": 94.5554,
110
- "eval_samples_per_second": 9.92,
111
- "eval_steps_per_second": 0.317,
112
- "step": 345
113
- },
114
- {
115
- "epoch": 5.072992700729927,
116
- "grad_norm": 15.120301246643066,
117
- "learning_rate": 2.492753623188406e-06,
118
- "loss": 0.18418676376342774,
119
- "step": 350
120
- },
121
- {
122
- "epoch": 5.802919708029197,
123
- "grad_norm": 20.06453514099121,
124
- "learning_rate": 2.847826086956522e-06,
125
- "loss": 0.16181354522705077,
126
- "step": 400
127
- },
128
- {
129
- "epoch": 6.0,
130
- "eval_accuracy": 0.9317697228144989,
131
- "eval_f1": 0.9571619812583668,
132
- "eval_loss": 0.17490684986114502,
133
- "eval_precision": 0.9675236806495264,
134
- "eval_recall": 0.9470198675496688,
135
- "eval_runtime": 78.099,
136
- "eval_samples_per_second": 12.01,
137
- "eval_steps_per_second": 0.384,
138
- "step": 414
139
- },
140
- {
141
- "epoch": 6.525547445255475,
142
- "grad_norm": 16.788759231567383,
143
- "learning_rate": 3.2101449275362324e-06,
144
- "loss": 0.15671510696411134,
145
- "step": 450
146
- },
147
- {
148
- "epoch": 7.0,
149
- "eval_accuracy": 0.9157782515991472,
150
- "eval_f1": 0.9460013670539986,
151
- "eval_loss": 0.22261112928390503,
152
- "eval_precision": 0.9774011299435028,
153
- "eval_recall": 0.9165562913907285,
154
- "eval_runtime": 92.5911,
155
- "eval_samples_per_second": 10.131,
156
- "eval_steps_per_second": 0.324,
157
- "step": 483
158
- },
159
- {
160
- "epoch": 7.248175182481752,
161
- "grad_norm": 36.52346420288086,
162
- "learning_rate": 3.5724637681159423e-06,
163
- "loss": 0.16417793273925782,
164
- "step": 500
165
- },
166
- {
167
- "epoch": 7.978102189781022,
168
- "grad_norm": 146.48744201660156,
169
- "learning_rate": 3.9347826086956525e-06,
170
- "loss": 0.13912958145141602,
171
- "step": 550
172
- },
173
- {
174
- "epoch": 8.0,
175
- "eval_accuracy": 0.9381663113006397,
176
- "eval_f1": 0.9617414248021108,
177
- "eval_loss": 0.15613189339637756,
178
- "eval_precision": 0.9579500657030223,
179
- "eval_recall": 0.9655629139072848,
180
- "eval_runtime": 125.8153,
181
- "eval_samples_per_second": 7.455,
182
- "eval_steps_per_second": 0.238,
183
- "step": 552
184
- },
185
- {
186
- "epoch": 8.700729927007298,
187
- "grad_norm": 37.7802734375,
188
- "learning_rate": 4.297101449275362e-06,
189
- "loss": 0.12975292205810546,
190
- "step": 600
191
- },
192
- {
193
- "epoch": 9.0,
194
- "eval_accuracy": 0.9402985074626866,
195
- "eval_f1": 0.9627659574468085,
196
- "eval_loss": 0.1664065420627594,
197
- "eval_precision": 0.9666221628838452,
198
- "eval_recall": 0.9589403973509933,
199
- "eval_runtime": 93.79,
200
- "eval_samples_per_second": 10.001,
201
- "eval_steps_per_second": 0.32,
202
- "step": 621
203
- },
204
- {
205
- "epoch": 9.423357664233576,
206
- "grad_norm": 39.47712707519531,
207
- "learning_rate": 4.659420289855073e-06,
208
- "loss": 0.12847793579101563,
209
- "step": 650
210
- },
211
- {
212
- "epoch": 10.0,
213
- "eval_accuracy": 0.9402985074626866,
214
- "eval_f1": 0.963302752293578,
215
- "eval_loss": 0.16533519327640533,
216
- "eval_precision": 0.953307392996109,
217
- "eval_recall": 0.9735099337748344,
218
- "eval_runtime": 92.125,
219
- "eval_samples_per_second": 10.182,
220
- "eval_steps_per_second": 0.326,
221
- "step": 690
222
- },
223
- {
224
- "epoch": 10.145985401459853,
225
- "grad_norm": 13.382551193237305,
226
- "learning_rate": 4.999997120818891e-06,
227
- "loss": 0.12533601760864257,
228
- "step": 700
229
- },
230
- {
231
- "epoch": 10.875912408759124,
232
- "grad_norm": 152.12037658691406,
233
- "learning_rate": 4.999101429246326e-06,
234
- "loss": 0.10754921913146973,
235
- "step": 750
236
- },
237
- {
238
- "epoch": 11.0,
239
- "eval_accuracy": 0.9402985074626866,
240
- "eval_f1": 0.9628154050464808,
241
- "eval_loss": 0.15896020829677582,
242
- "eval_precision": 0.9653794940079894,
243
- "eval_recall": 0.9602649006622517,
244
- "eval_runtime": 92.6027,
245
- "eval_samples_per_second": 10.129,
246
- "eval_steps_per_second": 0.324,
247
- "step": 759
248
- },
249
- {
250
- "epoch": 11.598540145985401,
251
- "grad_norm": 7.632987022399902,
252
- "learning_rate": 4.996606852480508e-06,
253
- "loss": 0.08507664680480957,
254
- "step": 800
255
- },
256
- {
257
- "epoch": 12.0,
258
- "eval_accuracy": 0.9275053304904051,
259
- "eval_f1": 0.9562419562419563,
260
- "eval_loss": 0.19671548902988434,
261
- "eval_precision": 0.9299123904881101,
262
- "eval_recall": 0.9841059602649007,
263
- "eval_runtime": 79.1855,
264
- "eval_samples_per_second": 11.846,
265
- "eval_steps_per_second": 0.379,
266
- "step": 828
267
- },
268
- {
269
- "epoch": 12.321167883211679,
270
- "grad_norm": 451.5773010253906,
271
- "learning_rate": 4.992514986511825e-06,
272
- "loss": 0.11187627792358398,
273
- "step": 850
274
- },
275
- {
276
- "epoch": 13.0,
277
- "eval_accuracy": 0.9339019189765458,
278
- "eval_f1": 0.9602053915275995,
279
- "eval_loss": 0.20483285188674927,
280
- "eval_precision": 0.9315068493150684,
281
- "eval_recall": 0.990728476821192,
282
- "eval_runtime": 84.5285,
283
- "eval_samples_per_second": 11.097,
284
- "eval_steps_per_second": 0.355,
285
- "step": 897
286
- },
287
- {
288
- "epoch": 13.043795620437956,
289
- "grad_norm": 27.037843704223633,
290
- "learning_rate": 4.986828449250792e-06,
291
- "loss": 0.08265207290649414,
292
- "step": 900
293
- },
294
- {
295
- "epoch": 13.773722627737227,
296
- "grad_norm": 58.419273376464844,
297
- "learning_rate": 4.979550878853154e-06,
298
- "loss": 0.08042881965637207,
299
- "step": 950
300
- },
301
- {
302
- "epoch": 14.0,
303
- "eval_accuracy": 0.9349680170575693,
304
- "eval_f1": 0.9605688429217841,
305
- "eval_loss": 0.21270374953746796,
306
- "eval_precision": 0.9381313131313131,
307
- "eval_recall": 0.9841059602649007,
308
- "eval_runtime": 92.7757,
309
- "eval_samples_per_second": 10.11,
310
- "eval_steps_per_second": 0.323,
311
- "step": 966
312
- },
313
- {
314
- "epoch": 14.496350364963504,
315
- "grad_norm": 60.74350357055664,
316
- "learning_rate": 4.97068693139226e-06,
317
- "loss": 0.08053727149963379,
318
- "step": 1000
319
- },
320
- {
321
- "epoch": 15.0,
322
- "eval_accuracy": 0.94136460554371,
323
- "eval_f1": 0.9635036496350365,
324
- "eval_loss": 0.18071264028549194,
325
- "eval_precision": 0.9654255319148937,
326
- "eval_recall": 0.9615894039735099,
327
- "eval_runtime": 87.3638,
328
- "eval_samples_per_second": 10.737,
329
- "eval_steps_per_second": 0.343,
330
- "step": 1035
331
- }
332
- ],
333
- "logging_steps": 50,
334
- "max_steps": 6900,
335
- "num_input_tokens_seen": 0,
336
- "num_train_epochs": 100,
337
- "save_steps": 500,
338
- "stateful_callbacks": {
339
- "TrainerControl": {
340
- "args": {
341
- "should_epoch_stop": false,
342
- "should_evaluate": false,
343
- "should_log": false,
344
- "should_save": true,
345
- "should_training_stop": false
346
- },
347
- "attributes": {}
348
- }
349
- },
350
- "total_flos": 1.80402939303552e+19,
351
- "train_batch_size": 16,
352
- "trial_name": null,
353
- "trial_params": null
354
- }