hwm21 commited on
Commit
e6714e4
1 Parent(s): 7aa2bee

Training in progress, step 6500, checkpoint

Browse files
checkpoint-6500/config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/detr-resnet-50-dc5",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "relu",
5
+ "architectures": [
6
+ "DetrForObjectDetection"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "auxiliary_loss": false,
10
+ "backbone": "resnet50",
11
+ "backbone_config": null,
12
+ "backbone_kwargs": {
13
+ "in_chans": 3,
14
+ "out_indices": [
15
+ 1,
16
+ 2,
17
+ 3,
18
+ 4
19
+ ],
20
+ "output_stride": 16
21
+ },
22
+ "bbox_cost": 5,
23
+ "bbox_loss_coefficient": 5,
24
+ "class_cost": 1,
25
+ "classifier_dropout": 0.0,
26
+ "d_model": 256,
27
+ "decoder_attention_heads": 8,
28
+ "decoder_ffn_dim": 2048,
29
+ "decoder_layerdrop": 0.0,
30
+ "decoder_layers": 6,
31
+ "dice_loss_coefficient": 1,
32
+ "dilation": true,
33
+ "dropout": 0.1,
34
+ "encoder_attention_heads": 8,
35
+ "encoder_ffn_dim": 2048,
36
+ "encoder_layerdrop": 0.0,
37
+ "encoder_layers": 6,
38
+ "eos_coefficient": 0.1,
39
+ "giou_cost": 2,
40
+ "giou_loss_coefficient": 2,
41
+ "id2label": {
42
+ "0": "segment"
43
+ },
44
+ "init_std": 0.02,
45
+ "init_xavier_std": 1.0,
46
+ "is_encoder_decoder": true,
47
+ "label2id": {
48
+ "segment": 0
49
+ },
50
+ "mask_loss_coefficient": 1,
51
+ "max_position_embeddings": 1024,
52
+ "model_type": "detr",
53
+ "num_channels": 3,
54
+ "num_hidden_layers": 6,
55
+ "num_queries": 500,
56
+ "position_embedding_type": "sine",
57
+ "scale_embedding": false,
58
+ "torch_dtype": "float32",
59
+ "transformers_version": "4.44.2",
60
+ "use_pretrained_backbone": true,
61
+ "use_timm_backbone": true
62
+ }
checkpoint-6500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3caa9e7cfce41d41c1c0daea28b53b7bf204d5d1630394c906d5fb01fbf85afe
3
+ size 166904424
checkpoint-6500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f19885a00be5a49d30232dd07d99733ff5c9b79daa905d161993769d94c6997c
3
+ size 331310970
checkpoint-6500/preprocessor_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_annotations": true,
3
+ "do_normalize": true,
4
+ "do_pad": true,
5
+ "do_rescale": true,
6
+ "do_resize": true,
7
+ "format": "coco_detection",
8
+ "image_mean": [
9
+ 0.485,
10
+ 0.456,
11
+ 0.406
12
+ ],
13
+ "image_processor_type": "DetrImageProcessor",
14
+ "image_std": [
15
+ 0.229,
16
+ 0.224,
17
+ 0.225
18
+ ],
19
+ "pad_size": null,
20
+ "resample": 2,
21
+ "rescale_factor": 0.00392156862745098,
22
+ "size": {
23
+ "longest_edge": 1333,
24
+ "shortest_edge": 800
25
+ }
26
+ }
checkpoint-6500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0302ee83b9c2a127c7752813e8c4a429e4f40cc35a65f1e5bcfced9f1091d3ba
3
+ size 14180
checkpoint-6500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b923daa357e4a43985affc206d7f59fa09b29cb28bd9d83c4518283dc362dfa
3
+ size 1064
checkpoint-6500/trainer_state.json ADDED
@@ -0,0 +1,488 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 100.0,
5
+ "eval_steps": 500,
6
+ "global_step": 6500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.5384615384615383,
13
+ "grad_norm": 134.7136688232422,
14
+ "learning_rate": 9.981600000000001e-06,
15
+ "loss": 5.6198,
16
+ "step": 100
17
+ },
18
+ {
19
+ "epoch": 3.076923076923077,
20
+ "grad_norm": 115.98611450195312,
21
+ "learning_rate": 9.961600000000001e-06,
22
+ "loss": 3.5959,
23
+ "step": 200
24
+ },
25
+ {
26
+ "epoch": 4.615384615384615,
27
+ "grad_norm": 114.23514556884766,
28
+ "learning_rate": 9.941600000000002e-06,
29
+ "loss": 3.3845,
30
+ "step": 300
31
+ },
32
+ {
33
+ "epoch": 6.153846153846154,
34
+ "grad_norm": 140.5648193359375,
35
+ "learning_rate": 9.921600000000002e-06,
36
+ "loss": 3.1494,
37
+ "step": 400
38
+ },
39
+ {
40
+ "epoch": 7.6923076923076925,
41
+ "grad_norm": 117.4485855102539,
42
+ "learning_rate": 9.901600000000002e-06,
43
+ "loss": 3.0773,
44
+ "step": 500
45
+ },
46
+ {
47
+ "epoch": 9.23076923076923,
48
+ "grad_norm": 389.71490478515625,
49
+ "learning_rate": 9.8816e-06,
50
+ "loss": 3.2116,
51
+ "step": 600
52
+ },
53
+ {
54
+ "epoch": 10.76923076923077,
55
+ "grad_norm": 107.11251831054688,
56
+ "learning_rate": 9.8616e-06,
57
+ "loss": 3.0471,
58
+ "step": 700
59
+ },
60
+ {
61
+ "epoch": 12.307692307692308,
62
+ "grad_norm": 85.49571228027344,
63
+ "learning_rate": 9.8416e-06,
64
+ "loss": 3.0201,
65
+ "step": 800
66
+ },
67
+ {
68
+ "epoch": 13.846153846153847,
69
+ "grad_norm": 121.62274932861328,
70
+ "learning_rate": 9.821600000000001e-06,
71
+ "loss": 2.9355,
72
+ "step": 900
73
+ },
74
+ {
75
+ "epoch": 15.384615384615385,
76
+ "grad_norm": 64.66451263427734,
77
+ "learning_rate": 9.801600000000001e-06,
78
+ "loss": 2.9637,
79
+ "step": 1000
80
+ },
81
+ {
82
+ "epoch": 16.923076923076923,
83
+ "grad_norm": 182.4657440185547,
84
+ "learning_rate": 9.781600000000001e-06,
85
+ "loss": 2.9819,
86
+ "step": 1100
87
+ },
88
+ {
89
+ "epoch": 18.46153846153846,
90
+ "grad_norm": 97.80529022216797,
91
+ "learning_rate": 9.761600000000002e-06,
92
+ "loss": 2.9486,
93
+ "step": 1200
94
+ },
95
+ {
96
+ "epoch": 20.0,
97
+ "grad_norm": 220.0562744140625,
98
+ "learning_rate": 9.741600000000002e-06,
99
+ "loss": 2.8608,
100
+ "step": 1300
101
+ },
102
+ {
103
+ "epoch": 21.53846153846154,
104
+ "grad_norm": 95.53397369384766,
105
+ "learning_rate": 9.7216e-06,
106
+ "loss": 2.8322,
107
+ "step": 1400
108
+ },
109
+ {
110
+ "epoch": 23.076923076923077,
111
+ "grad_norm": 67.54853057861328,
112
+ "learning_rate": 9.7016e-06,
113
+ "loss": 2.9429,
114
+ "step": 1500
115
+ },
116
+ {
117
+ "epoch": 24.615384615384617,
118
+ "grad_norm": 214.53131103515625,
119
+ "learning_rate": 9.6816e-06,
120
+ "loss": 2.7927,
121
+ "step": 1600
122
+ },
123
+ {
124
+ "epoch": 26.153846153846153,
125
+ "grad_norm": 293.3318786621094,
126
+ "learning_rate": 9.6616e-06,
127
+ "loss": 2.7665,
128
+ "step": 1700
129
+ },
130
+ {
131
+ "epoch": 27.692307692307693,
132
+ "grad_norm": 216.3682861328125,
133
+ "learning_rate": 9.641600000000001e-06,
134
+ "loss": 2.8309,
135
+ "step": 1800
136
+ },
137
+ {
138
+ "epoch": 29.23076923076923,
139
+ "grad_norm": 168.0605010986328,
140
+ "learning_rate": 9.621600000000001e-06,
141
+ "loss": 2.8433,
142
+ "step": 1900
143
+ },
144
+ {
145
+ "epoch": 30.76923076923077,
146
+ "grad_norm": 103.49143981933594,
147
+ "learning_rate": 9.601600000000001e-06,
148
+ "loss": 2.7606,
149
+ "step": 2000
150
+ },
151
+ {
152
+ "epoch": 32.30769230769231,
153
+ "grad_norm": 116.6761474609375,
154
+ "learning_rate": 9.581600000000002e-06,
155
+ "loss": 2.6595,
156
+ "step": 2100
157
+ },
158
+ {
159
+ "epoch": 33.84615384615385,
160
+ "grad_norm": 176.0087432861328,
161
+ "learning_rate": 9.5616e-06,
162
+ "loss": 2.632,
163
+ "step": 2200
164
+ },
165
+ {
166
+ "epoch": 35.38461538461539,
167
+ "grad_norm": 120.39679718017578,
168
+ "learning_rate": 9.5416e-06,
169
+ "loss": 2.5941,
170
+ "step": 2300
171
+ },
172
+ {
173
+ "epoch": 36.92307692307692,
174
+ "grad_norm": 100.69256591796875,
175
+ "learning_rate": 9.5216e-06,
176
+ "loss": 2.6007,
177
+ "step": 2400
178
+ },
179
+ {
180
+ "epoch": 38.46153846153846,
181
+ "grad_norm": 254.54440307617188,
182
+ "learning_rate": 9.5016e-06,
183
+ "loss": 2.5548,
184
+ "step": 2500
185
+ },
186
+ {
187
+ "epoch": 40.0,
188
+ "grad_norm": 113.618896484375,
189
+ "learning_rate": 9.4816e-06,
190
+ "loss": 2.5157,
191
+ "step": 2600
192
+ },
193
+ {
194
+ "epoch": 41.53846153846154,
195
+ "grad_norm": 825.5616455078125,
196
+ "learning_rate": 9.461600000000001e-06,
197
+ "loss": 2.4832,
198
+ "step": 2700
199
+ },
200
+ {
201
+ "epoch": 43.07692307692308,
202
+ "grad_norm": 58.83940505981445,
203
+ "learning_rate": 9.441600000000001e-06,
204
+ "loss": 2.4393,
205
+ "step": 2800
206
+ },
207
+ {
208
+ "epoch": 44.61538461538461,
209
+ "grad_norm": 89.66182708740234,
210
+ "learning_rate": 9.421600000000001e-06,
211
+ "loss": 2.4381,
212
+ "step": 2900
213
+ },
214
+ {
215
+ "epoch": 46.15384615384615,
216
+ "grad_norm": 242.81410217285156,
217
+ "learning_rate": 9.4016e-06,
218
+ "loss": 2.3978,
219
+ "step": 3000
220
+ },
221
+ {
222
+ "epoch": 47.69230769230769,
223
+ "grad_norm": 70.43364715576172,
224
+ "learning_rate": 9.3816e-06,
225
+ "loss": 2.3782,
226
+ "step": 3100
227
+ },
228
+ {
229
+ "epoch": 49.23076923076923,
230
+ "grad_norm": 244.31356811523438,
231
+ "learning_rate": 9.3616e-06,
232
+ "loss": 2.3319,
233
+ "step": 3200
234
+ },
235
+ {
236
+ "epoch": 50.76923076923077,
237
+ "grad_norm": 59.848697662353516,
238
+ "learning_rate": 9.3416e-06,
239
+ "loss": 2.3205,
240
+ "step": 3300
241
+ },
242
+ {
243
+ "epoch": 52.30769230769231,
244
+ "grad_norm": 109.26374053955078,
245
+ "learning_rate": 9.3216e-06,
246
+ "loss": 2.2943,
247
+ "step": 3400
248
+ },
249
+ {
250
+ "epoch": 53.84615384615385,
251
+ "grad_norm": 307.2849426269531,
252
+ "learning_rate": 9.301600000000001e-06,
253
+ "loss": 2.2913,
254
+ "step": 3500
255
+ },
256
+ {
257
+ "epoch": 55.38461538461539,
258
+ "grad_norm": 69.93729400634766,
259
+ "learning_rate": 9.281600000000001e-06,
260
+ "loss": 2.2166,
261
+ "step": 3600
262
+ },
263
+ {
264
+ "epoch": 56.92307692307692,
265
+ "grad_norm": 83.3631591796875,
266
+ "learning_rate": 9.261600000000001e-06,
267
+ "loss": 2.2718,
268
+ "step": 3700
269
+ },
270
+ {
271
+ "epoch": 58.46153846153846,
272
+ "grad_norm": 48.52922821044922,
273
+ "learning_rate": 9.2416e-06,
274
+ "loss": 2.2645,
275
+ "step": 3800
276
+ },
277
+ {
278
+ "epoch": 60.0,
279
+ "grad_norm": 51.951324462890625,
280
+ "learning_rate": 9.2218e-06,
281
+ "loss": 2.2416,
282
+ "step": 3900
283
+ },
284
+ {
285
+ "epoch": 61.53846153846154,
286
+ "grad_norm": 254.99429321289062,
287
+ "learning_rate": 9.2018e-06,
288
+ "loss": 2.2104,
289
+ "step": 4000
290
+ },
291
+ {
292
+ "epoch": 63.07692307692308,
293
+ "grad_norm": 109.68157958984375,
294
+ "learning_rate": 9.1818e-06,
295
+ "loss": 2.1574,
296
+ "step": 4100
297
+ },
298
+ {
299
+ "epoch": 64.61538461538461,
300
+ "grad_norm": 85.5855941772461,
301
+ "learning_rate": 9.161800000000001e-06,
302
+ "loss": 2.1513,
303
+ "step": 4200
304
+ },
305
+ {
306
+ "epoch": 66.15384615384616,
307
+ "grad_norm": 63.89781951904297,
308
+ "learning_rate": 9.141800000000001e-06,
309
+ "loss": 2.1331,
310
+ "step": 4300
311
+ },
312
+ {
313
+ "epoch": 67.6923076923077,
314
+ "grad_norm": 184.29612731933594,
315
+ "learning_rate": 9.121800000000001e-06,
316
+ "loss": 2.1866,
317
+ "step": 4400
318
+ },
319
+ {
320
+ "epoch": 69.23076923076923,
321
+ "grad_norm": 108.59600830078125,
322
+ "learning_rate": 9.1018e-06,
323
+ "loss": 2.1453,
324
+ "step": 4500
325
+ },
326
+ {
327
+ "epoch": 70.76923076923077,
328
+ "grad_norm": 80.51258087158203,
329
+ "learning_rate": 9.0818e-06,
330
+ "loss": 2.1125,
331
+ "step": 4600
332
+ },
333
+ {
334
+ "epoch": 72.3076923076923,
335
+ "grad_norm": 172.7328643798828,
336
+ "learning_rate": 9.0618e-06,
337
+ "loss": 2.141,
338
+ "step": 4700
339
+ },
340
+ {
341
+ "epoch": 73.84615384615384,
342
+ "grad_norm": 58.20785903930664,
343
+ "learning_rate": 9.0418e-06,
344
+ "loss": 2.0786,
345
+ "step": 4800
346
+ },
347
+ {
348
+ "epoch": 75.38461538461539,
349
+ "grad_norm": 72.46143341064453,
350
+ "learning_rate": 9.0218e-06,
351
+ "loss": 2.0914,
352
+ "step": 4900
353
+ },
354
+ {
355
+ "epoch": 76.92307692307692,
356
+ "grad_norm": 48.18838119506836,
357
+ "learning_rate": 9.0018e-06,
358
+ "loss": 2.0439,
359
+ "step": 5000
360
+ },
361
+ {
362
+ "epoch": 78.46153846153847,
363
+ "grad_norm": 87.95365905761719,
364
+ "learning_rate": 8.981800000000001e-06,
365
+ "loss": 2.0331,
366
+ "step": 5100
367
+ },
368
+ {
369
+ "epoch": 80.0,
370
+ "grad_norm": 99.82858276367188,
371
+ "learning_rate": 8.961800000000001e-06,
372
+ "loss": 2.1268,
373
+ "step": 5200
374
+ },
375
+ {
376
+ "epoch": 81.53846153846153,
377
+ "grad_norm": 85.33229064941406,
378
+ "learning_rate": 8.9418e-06,
379
+ "loss": 2.028,
380
+ "step": 5300
381
+ },
382
+ {
383
+ "epoch": 83.07692307692308,
384
+ "grad_norm": 83.01776885986328,
385
+ "learning_rate": 8.9218e-06,
386
+ "loss": 2.0179,
387
+ "step": 5400
388
+ },
389
+ {
390
+ "epoch": 84.61538461538461,
391
+ "grad_norm": 95.0901107788086,
392
+ "learning_rate": 8.9018e-06,
393
+ "loss": 2.0825,
394
+ "step": 5500
395
+ },
396
+ {
397
+ "epoch": 86.15384615384616,
398
+ "grad_norm": 85.66802978515625,
399
+ "learning_rate": 8.8818e-06,
400
+ "loss": 2.0588,
401
+ "step": 5600
402
+ },
403
+ {
404
+ "epoch": 87.6923076923077,
405
+ "grad_norm": 69.63407897949219,
406
+ "learning_rate": 8.8618e-06,
407
+ "loss": 2.0298,
408
+ "step": 5700
409
+ },
410
+ {
411
+ "epoch": 89.23076923076923,
412
+ "grad_norm": 174.88063049316406,
413
+ "learning_rate": 8.8418e-06,
414
+ "loss": 1.9867,
415
+ "step": 5800
416
+ },
417
+ {
418
+ "epoch": 90.76923076923077,
419
+ "grad_norm": 65.64617156982422,
420
+ "learning_rate": 8.8218e-06,
421
+ "loss": 2.0674,
422
+ "step": 5900
423
+ },
424
+ {
425
+ "epoch": 92.3076923076923,
426
+ "grad_norm": 152.2218780517578,
427
+ "learning_rate": 8.802e-06,
428
+ "loss": 1.992,
429
+ "step": 6000
430
+ },
431
+ {
432
+ "epoch": 93.84615384615384,
433
+ "grad_norm": 74.4027328491211,
434
+ "learning_rate": 8.782e-06,
435
+ "loss": 2.0016,
436
+ "step": 6100
437
+ },
438
+ {
439
+ "epoch": 95.38461538461539,
440
+ "grad_norm": 67.4207992553711,
441
+ "learning_rate": 8.762e-06,
442
+ "loss": 2.0001,
443
+ "step": 6200
444
+ },
445
+ {
446
+ "epoch": 96.92307692307692,
447
+ "grad_norm": 182.99061584472656,
448
+ "learning_rate": 8.742e-06,
449
+ "loss": 1.9897,
450
+ "step": 6300
451
+ },
452
+ {
453
+ "epoch": 98.46153846153847,
454
+ "grad_norm": 70.14286804199219,
455
+ "learning_rate": 8.722e-06,
456
+ "loss": 2.0146,
457
+ "step": 6400
458
+ },
459
+ {
460
+ "epoch": 100.0,
461
+ "grad_norm": 117.57647705078125,
462
+ "learning_rate": 8.702e-06,
463
+ "loss": 1.9298,
464
+ "step": 6500
465
+ }
466
+ ],
467
+ "logging_steps": 100,
468
+ "max_steps": 50000,
469
+ "num_input_tokens_seen": 0,
470
+ "num_train_epochs": 770,
471
+ "save_steps": 500,
472
+ "stateful_callbacks": {
473
+ "TrainerControl": {
474
+ "args": {
475
+ "should_epoch_stop": false,
476
+ "should_evaluate": false,
477
+ "should_log": false,
478
+ "should_save": true,
479
+ "should_training_stop": false
480
+ },
481
+ "attributes": {}
482
+ }
483
+ },
484
+ "total_flos": 2.484580580352e+19,
485
+ "train_batch_size": 8,
486
+ "trial_name": null,
487
+ "trial_params": null
488
+ }
checkpoint-6500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee77453dae7076acf919eb7bcd8ccb5a082573a80b4f2c85a6a7d7ad1edcbedf
3
+ size 5240