sam1120 commited on
Commit
6c4136e
1 Parent(s): 9d2eeb4

Training in progress, step 80

Browse files
checkpoint-80/config.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "nvidia/mit-b0",
3
+ "architectures": [
4
+ "SegformerForSemanticSegmentation"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.0,
7
+ "classifier_dropout_prob": 0.1,
8
+ "decoder_hidden_size": 256,
9
+ "depths": [
10
+ 2,
11
+ 2,
12
+ 2,
13
+ 2
14
+ ],
15
+ "downsampling_rates": [
16
+ 1,
17
+ 4,
18
+ 8,
19
+ 16
20
+ ],
21
+ "drop_path_rate": 0.1,
22
+ "hidden_act": "gelu",
23
+ "hidden_dropout_prob": 0.0,
24
+ "hidden_sizes": [
25
+ 32,
26
+ 64,
27
+ 160,
28
+ 256
29
+ ],
30
+ "id2label": {
31
+ "0": "unlabeled",
32
+ "1": "dropoff",
33
+ "2": "undropoff"
34
+ },
35
+ "image_size": 224,
36
+ "initializer_range": 0.02,
37
+ "label2id": {
38
+ "dropoff": 1,
39
+ "undropoff": 2,
40
+ "unlabeled": 0
41
+ },
42
+ "layer_norm_eps": 1e-06,
43
+ "mlp_ratios": [
44
+ 4,
45
+ 4,
46
+ 4,
47
+ 4
48
+ ],
49
+ "model_type": "segformer",
50
+ "num_attention_heads": [
51
+ 1,
52
+ 2,
53
+ 5,
54
+ 8
55
+ ],
56
+ "num_channels": 3,
57
+ "num_encoder_blocks": 4,
58
+ "patch_sizes": [
59
+ 7,
60
+ 3,
61
+ 3,
62
+ 3
63
+ ],
64
+ "reshape_last_stage": true,
65
+ "semantic_loss_ignore_index": 255,
66
+ "sr_ratios": [
67
+ 8,
68
+ 4,
69
+ 2,
70
+ 1
71
+ ],
72
+ "strides": [
73
+ 4,
74
+ 2,
75
+ 2,
76
+ 2
77
+ ],
78
+ "torch_dtype": "float32",
79
+ "transformers_version": "4.30.2"
80
+ }
checkpoint-80/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e59a2322d6cfe36c9577fdacf737149c270419619b58612fdc94c7dc2ddbc5b2
3
+ size 29838853
checkpoint-80/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf3c492380c1369ce8f366fb3614bd6658de30c88484413eee29dd205e35e466
3
+ size 14932813
checkpoint-80/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:246bd42a635d0a5a548ccbc5aadc2869a4ce5c5c3b99a959193d271da4847b33
3
+ size 14575
checkpoint-80/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35a7512f60c09f2600e8cbb580a1637eb278b2d90111b83bcc03463020b9e0ad
3
+ size 627
checkpoint-80/trainer_state.json ADDED
@@ -0,0 +1,632 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.3731442093849182,
3
+ "best_model_checkpoint": "/robodata/smodak/Projects/nspl/scripts/terrainseg/training/models/dropoff-utcustom-train-SF-RGB-b0_6/checkpoint-80",
4
+ "epoch": 26.666666666666668,
5
+ "global_step": 80,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.33,
12
+ "learning_rate": 3.888888888888889e-06,
13
+ "loss": 1.1755,
14
+ "step": 1
15
+ },
16
+ {
17
+ "epoch": 0.67,
18
+ "learning_rate": 7.777777777777777e-06,
19
+ "loss": 1.1749,
20
+ "step": 2
21
+ },
22
+ {
23
+ "epoch": 1.0,
24
+ "learning_rate": 1.1666666666666665e-05,
25
+ "loss": 1.183,
26
+ "step": 3
27
+ },
28
+ {
29
+ "epoch": 1.33,
30
+ "learning_rate": 1.5555555555555555e-05,
31
+ "loss": 1.1718,
32
+ "step": 4
33
+ },
34
+ {
35
+ "epoch": 1.67,
36
+ "learning_rate": 1.9444444444444442e-05,
37
+ "loss": 1.1706,
38
+ "step": 5
39
+ },
40
+ {
41
+ "epoch": 2.0,
42
+ "learning_rate": 2.333333333333333e-05,
43
+ "loss": 1.1696,
44
+ "step": 6
45
+ },
46
+ {
47
+ "epoch": 2.33,
48
+ "learning_rate": 2.722222222222222e-05,
49
+ "loss": 1.1558,
50
+ "step": 7
51
+ },
52
+ {
53
+ "epoch": 2.67,
54
+ "learning_rate": 3.111111111111111e-05,
55
+ "loss": 1.1489,
56
+ "step": 8
57
+ },
58
+ {
59
+ "epoch": 3.0,
60
+ "learning_rate": 3.5e-05,
61
+ "loss": 1.135,
62
+ "step": 9
63
+ },
64
+ {
65
+ "epoch": 3.33,
66
+ "learning_rate": 3.8888888888888884e-05,
67
+ "loss": 1.1234,
68
+ "step": 10
69
+ },
70
+ {
71
+ "epoch": 3.33,
72
+ "eval_accuracy_dropoff": 0.772015071105291,
73
+ "eval_accuracy_undropoff": 0.353825865102452,
74
+ "eval_accuracy_unlabeled": NaN,
75
+ "eval_iou_dropoff": 0.1801195043477445,
76
+ "eval_iou_undropoff": 0.3536296682807036,
77
+ "eval_iou_unlabeled": 0.0,
78
+ "eval_loss": 1.097259283065796,
79
+ "eval_mean_accuracy": 0.5629204681038715,
80
+ "eval_mean_iou": 0.17791639087614938,
81
+ "eval_overall_accuracy": 0.37234242757161456,
82
+ "eval_runtime": 1.873,
83
+ "eval_samples_per_second": 8.008,
84
+ "eval_steps_per_second": 0.534,
85
+ "step": 10
86
+ },
87
+ {
88
+ "epoch": 3.67,
89
+ "learning_rate": 4.277777777777778e-05,
90
+ "loss": 1.1168,
91
+ "step": 11
92
+ },
93
+ {
94
+ "epoch": 4.0,
95
+ "learning_rate": 4.666666666666666e-05,
96
+ "loss": 1.0895,
97
+ "step": 12
98
+ },
99
+ {
100
+ "epoch": 4.33,
101
+ "learning_rate": 5.055555555555555e-05,
102
+ "loss": 1.1027,
103
+ "step": 13
104
+ },
105
+ {
106
+ "epoch": 4.67,
107
+ "learning_rate": 5.444444444444444e-05,
108
+ "loss": 1.0795,
109
+ "step": 14
110
+ },
111
+ {
112
+ "epoch": 5.0,
113
+ "learning_rate": 5.833333333333333e-05,
114
+ "loss": 1.0598,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 5.33,
119
+ "learning_rate": 6.222222222222222e-05,
120
+ "loss": 1.0449,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 5.67,
125
+ "learning_rate": 6.611111111111111e-05,
126
+ "loss": 1.0268,
127
+ "step": 17
128
+ },
129
+ {
130
+ "epoch": 6.0,
131
+ "learning_rate": 7e-05,
132
+ "loss": 1.0843,
133
+ "step": 18
134
+ },
135
+ {
136
+ "epoch": 6.33,
137
+ "learning_rate": 6.97953216374269e-05,
138
+ "loss": 1.0029,
139
+ "step": 19
140
+ },
141
+ {
142
+ "epoch": 6.67,
143
+ "learning_rate": 6.95906432748538e-05,
144
+ "loss": 0.975,
145
+ "step": 20
146
+ },
147
+ {
148
+ "epoch": 6.67,
149
+ "eval_accuracy_dropoff": 0.8258609598639924,
150
+ "eval_accuracy_undropoff": 0.8102080545984994,
151
+ "eval_accuracy_unlabeled": NaN,
152
+ "eval_iou_dropoff": 0.24279987572018127,
153
+ "eval_iou_undropoff": 0.8068537963656117,
154
+ "eval_iou_unlabeled": 0.0,
155
+ "eval_loss": 1.0260422229766846,
156
+ "eval_mean_accuracy": 0.8180345072312459,
157
+ "eval_mean_iou": 0.34988455736193097,
158
+ "eval_overall_accuracy": 0.810901133219401,
159
+ "eval_runtime": 1.6566,
160
+ "eval_samples_per_second": 9.055,
161
+ "eval_steps_per_second": 0.604,
162
+ "step": 20
163
+ },
164
+ {
165
+ "epoch": 7.0,
166
+ "learning_rate": 6.938596491228069e-05,
167
+ "loss": 1.0358,
168
+ "step": 21
169
+ },
170
+ {
171
+ "epoch": 7.33,
172
+ "learning_rate": 6.91812865497076e-05,
173
+ "loss": 0.9361,
174
+ "step": 22
175
+ },
176
+ {
177
+ "epoch": 7.67,
178
+ "learning_rate": 6.89766081871345e-05,
179
+ "loss": 0.9311,
180
+ "step": 23
181
+ },
182
+ {
183
+ "epoch": 8.0,
184
+ "learning_rate": 6.877192982456139e-05,
185
+ "loss": 0.9149,
186
+ "step": 24
187
+ },
188
+ {
189
+ "epoch": 8.33,
190
+ "learning_rate": 6.85672514619883e-05,
191
+ "loss": 0.9056,
192
+ "step": 25
193
+ },
194
+ {
195
+ "epoch": 8.67,
196
+ "learning_rate": 6.83625730994152e-05,
197
+ "loss": 0.8637,
198
+ "step": 26
199
+ },
200
+ {
201
+ "epoch": 9.0,
202
+ "learning_rate": 6.81578947368421e-05,
203
+ "loss": 0.7991,
204
+ "step": 27
205
+ },
206
+ {
207
+ "epoch": 9.33,
208
+ "learning_rate": 6.7953216374269e-05,
209
+ "loss": 0.8143,
210
+ "step": 28
211
+ },
212
+ {
213
+ "epoch": 9.67,
214
+ "learning_rate": 6.774853801169589e-05,
215
+ "loss": 0.7999,
216
+ "step": 29
217
+ },
218
+ {
219
+ "epoch": 10.0,
220
+ "learning_rate": 6.75438596491228e-05,
221
+ "loss": 0.9464,
222
+ "step": 30
223
+ },
224
+ {
225
+ "epoch": 10.0,
226
+ "eval_accuracy_dropoff": 0.5212052289383601,
227
+ "eval_accuracy_undropoff": 0.9700488444545206,
228
+ "eval_accuracy_unlabeled": NaN,
229
+ "eval_iou_dropoff": 0.3384996437670423,
230
+ "eval_iou_undropoff": 0.9507046257182781,
231
+ "eval_iou_unlabeled": 0.0,
232
+ "eval_loss": 0.8130465149879456,
233
+ "eval_mean_accuracy": 0.7456270366964404,
234
+ "eval_mean_iou": 0.4297347564951068,
235
+ "eval_overall_accuracy": 0.9501749674479166,
236
+ "eval_runtime": 1.7097,
237
+ "eval_samples_per_second": 8.773,
238
+ "eval_steps_per_second": 0.585,
239
+ "step": 30
240
+ },
241
+ {
242
+ "epoch": 10.33,
243
+ "learning_rate": 6.733918128654971e-05,
244
+ "loss": 0.763,
245
+ "step": 31
246
+ },
247
+ {
248
+ "epoch": 10.67,
249
+ "learning_rate": 6.71345029239766e-05,
250
+ "loss": 0.7591,
251
+ "step": 32
252
+ },
253
+ {
254
+ "epoch": 11.0,
255
+ "learning_rate": 6.69298245614035e-05,
256
+ "loss": 0.6623,
257
+ "step": 33
258
+ },
259
+ {
260
+ "epoch": 11.33,
261
+ "learning_rate": 6.672514619883041e-05,
262
+ "loss": 0.7138,
263
+ "step": 34
264
+ },
265
+ {
266
+ "epoch": 11.67,
267
+ "learning_rate": 6.65204678362573e-05,
268
+ "loss": 0.6695,
269
+ "step": 35
270
+ },
271
+ {
272
+ "epoch": 12.0,
273
+ "learning_rate": 6.63157894736842e-05,
274
+ "loss": 0.6987,
275
+ "step": 36
276
+ },
277
+ {
278
+ "epoch": 12.33,
279
+ "learning_rate": 6.611111111111111e-05,
280
+ "loss": 0.6494,
281
+ "step": 37
282
+ },
283
+ {
284
+ "epoch": 12.67,
285
+ "learning_rate": 6.5906432748538e-05,
286
+ "loss": 0.6317,
287
+ "step": 38
288
+ },
289
+ {
290
+ "epoch": 13.0,
291
+ "learning_rate": 6.570175438596491e-05,
292
+ "loss": 0.834,
293
+ "step": 39
294
+ },
295
+ {
296
+ "epoch": 13.33,
297
+ "learning_rate": 6.54970760233918e-05,
298
+ "loss": 0.6167,
299
+ "step": 40
300
+ },
301
+ {
302
+ "epoch": 13.33,
303
+ "eval_accuracy_dropoff": 0.5047671560181037,
304
+ "eval_accuracy_undropoff": 0.9829004494881923,
305
+ "eval_accuracy_unlabeled": NaN,
306
+ "eval_iou_dropoff": 0.3743041743152479,
307
+ "eval_iou_undropoff": 0.961015278770039,
308
+ "eval_iou_unlabeled": 0.0,
309
+ "eval_loss": 0.6001297235488892,
310
+ "eval_mean_accuracy": 0.743833802753148,
311
+ "eval_mean_iou": 0.44510648436176226,
312
+ "eval_overall_accuracy": 0.9617296854654948,
313
+ "eval_runtime": 1.7423,
314
+ "eval_samples_per_second": 8.609,
315
+ "eval_steps_per_second": 0.574,
316
+ "step": 40
317
+ },
318
+ {
319
+ "epoch": 13.67,
320
+ "learning_rate": 6.52923976608187e-05,
321
+ "loss": 0.6201,
322
+ "step": 41
323
+ },
324
+ {
325
+ "epoch": 14.0,
326
+ "learning_rate": 6.508771929824561e-05,
327
+ "loss": 0.6009,
328
+ "step": 42
329
+ },
330
+ {
331
+ "epoch": 14.33,
332
+ "learning_rate": 6.488304093567252e-05,
333
+ "loss": 0.5716,
334
+ "step": 43
335
+ },
336
+ {
337
+ "epoch": 14.67,
338
+ "learning_rate": 6.467836257309941e-05,
339
+ "loss": 0.5779,
340
+ "step": 44
341
+ },
342
+ {
343
+ "epoch": 15.0,
344
+ "learning_rate": 6.447368421052631e-05,
345
+ "loss": 0.5397,
346
+ "step": 45
347
+ },
348
+ {
349
+ "epoch": 15.33,
350
+ "learning_rate": 6.426900584795322e-05,
351
+ "loss": 0.5569,
352
+ "step": 46
353
+ },
354
+ {
355
+ "epoch": 15.67,
356
+ "learning_rate": 6.406432748538011e-05,
357
+ "loss": 0.5152,
358
+ "step": 47
359
+ },
360
+ {
361
+ "epoch": 16.0,
362
+ "learning_rate": 6.3859649122807e-05,
363
+ "loss": 0.5266,
364
+ "step": 48
365
+ },
366
+ {
367
+ "epoch": 16.33,
368
+ "learning_rate": 6.365497076023391e-05,
369
+ "loss": 0.5389,
370
+ "step": 49
371
+ },
372
+ {
373
+ "epoch": 16.67,
374
+ "learning_rate": 6.345029239766081e-05,
375
+ "loss": 0.4818,
376
+ "step": 50
377
+ },
378
+ {
379
+ "epoch": 16.67,
380
+ "eval_accuracy_dropoff": 0.4572736462425621,
381
+ "eval_accuracy_undropoff": 0.9901589972677334,
382
+ "eval_accuracy_unlabeled": NaN,
383
+ "eval_iou_dropoff": 0.38151532722193204,
384
+ "eval_iou_undropoff": 0.9659037872797944,
385
+ "eval_iou_unlabeled": 0.0,
386
+ "eval_loss": 0.46292775869369507,
387
+ "eval_mean_accuracy": 0.7237163217551478,
388
+ "eval_mean_iou": 0.4491397048339088,
389
+ "eval_overall_accuracy": 0.9665639241536458,
390
+ "eval_runtime": 1.7358,
391
+ "eval_samples_per_second": 8.641,
392
+ "eval_steps_per_second": 0.576,
393
+ "step": 50
394
+ },
395
+ {
396
+ "epoch": 17.0,
397
+ "learning_rate": 6.324561403508772e-05,
398
+ "loss": 0.5289,
399
+ "step": 51
400
+ },
401
+ {
402
+ "epoch": 17.33,
403
+ "learning_rate": 6.304093567251461e-05,
404
+ "loss": 0.4991,
405
+ "step": 52
406
+ },
407
+ {
408
+ "epoch": 17.67,
409
+ "learning_rate": 6.283625730994151e-05,
410
+ "loss": 0.4643,
411
+ "step": 53
412
+ },
413
+ {
414
+ "epoch": 18.0,
415
+ "learning_rate": 6.263157894736842e-05,
416
+ "loss": 0.4359,
417
+ "step": 54
418
+ },
419
+ {
420
+ "epoch": 18.33,
421
+ "learning_rate": 6.242690058479532e-05,
422
+ "loss": 0.4679,
423
+ "step": 55
424
+ },
425
+ {
426
+ "epoch": 18.67,
427
+ "learning_rate": 6.222222222222222e-05,
428
+ "loss": 0.4619,
429
+ "step": 56
430
+ },
431
+ {
432
+ "epoch": 19.0,
433
+ "learning_rate": 6.201754385964911e-05,
434
+ "loss": 0.6641,
435
+ "step": 57
436
+ },
437
+ {
438
+ "epoch": 19.33,
439
+ "learning_rate": 6.181286549707602e-05,
440
+ "loss": 0.4316,
441
+ "step": 58
442
+ },
443
+ {
444
+ "epoch": 19.67,
445
+ "learning_rate": 6.160818713450292e-05,
446
+ "loss": 0.4475,
447
+ "step": 59
448
+ },
449
+ {
450
+ "epoch": 20.0,
451
+ "learning_rate": 6.140350877192981e-05,
452
+ "loss": 0.4733,
453
+ "step": 60
454
+ },
455
+ {
456
+ "epoch": 20.0,
457
+ "eval_accuracy_dropoff": 0.42558641762584143,
458
+ "eval_accuracy_undropoff": 0.987872440296196,
459
+ "eval_accuracy_unlabeled": NaN,
460
+ "eval_iou_dropoff": 0.33826825716385683,
461
+ "eval_iou_undropoff": 0.9622649725509712,
462
+ "eval_iou_unlabeled": 0.0,
463
+ "eval_loss": 0.43790048360824585,
464
+ "eval_mean_accuracy": 0.7067294289610188,
465
+ "eval_mean_iou": 0.4335110765716093,
466
+ "eval_overall_accuracy": 0.9629755655924479,
467
+ "eval_runtime": 1.7742,
468
+ "eval_samples_per_second": 8.455,
469
+ "eval_steps_per_second": 0.564,
470
+ "step": 60
471
+ },
472
+ {
473
+ "epoch": 20.33,
474
+ "learning_rate": 6.119883040935672e-05,
475
+ "loss": 0.423,
476
+ "step": 61
477
+ },
478
+ {
479
+ "epoch": 20.67,
480
+ "learning_rate": 6.0994152046783624e-05,
481
+ "loss": 0.4219,
482
+ "step": 62
483
+ },
484
+ {
485
+ "epoch": 21.0,
486
+ "learning_rate": 6.0789473684210525e-05,
487
+ "loss": 0.6326,
488
+ "step": 63
489
+ },
490
+ {
491
+ "epoch": 21.33,
492
+ "learning_rate": 6.058479532163742e-05,
493
+ "loss": 0.4299,
494
+ "step": 64
495
+ },
496
+ {
497
+ "epoch": 21.67,
498
+ "learning_rate": 6.038011695906432e-05,
499
+ "loss": 0.4147,
500
+ "step": 65
501
+ },
502
+ {
503
+ "epoch": 22.0,
504
+ "learning_rate": 6.0175438596491224e-05,
505
+ "loss": 0.4815,
506
+ "step": 66
507
+ },
508
+ {
509
+ "epoch": 22.33,
510
+ "learning_rate": 5.9970760233918126e-05,
511
+ "loss": 0.4477,
512
+ "step": 67
513
+ },
514
+ {
515
+ "epoch": 22.67,
516
+ "learning_rate": 5.976608187134502e-05,
517
+ "loss": 0.4031,
518
+ "step": 68
519
+ },
520
+ {
521
+ "epoch": 23.0,
522
+ "learning_rate": 5.956140350877192e-05,
523
+ "loss": 0.6048,
524
+ "step": 69
525
+ },
526
+ {
527
+ "epoch": 23.33,
528
+ "learning_rate": 5.9356725146198824e-05,
529
+ "loss": 0.3843,
530
+ "step": 70
531
+ },
532
+ {
533
+ "epoch": 23.33,
534
+ "eval_accuracy_dropoff": 0.3821076573161486,
535
+ "eval_accuracy_undropoff": 0.992239862567096,
536
+ "eval_accuracy_unlabeled": NaN,
537
+ "eval_iou_dropoff": 0.328346864743479,
538
+ "eval_iou_undropoff": 0.9646259654307497,
539
+ "eval_iou_unlabeled": 0.0,
540
+ "eval_loss": 0.4072829484939575,
541
+ "eval_mean_accuracy": 0.6871737599416223,
542
+ "eval_mean_iou": 0.4309909433914096,
543
+ "eval_overall_accuracy": 0.9652244567871093,
544
+ "eval_runtime": 1.8122,
545
+ "eval_samples_per_second": 8.277,
546
+ "eval_steps_per_second": 0.552,
547
+ "step": 70
548
+ },
549
+ {
550
+ "epoch": 23.67,
551
+ "learning_rate": 5.9152046783625726e-05,
552
+ "loss": 0.3967,
553
+ "step": 71
554
+ },
555
+ {
556
+ "epoch": 24.0,
557
+ "learning_rate": 5.894736842105262e-05,
558
+ "loss": 0.4051,
559
+ "step": 72
560
+ },
561
+ {
562
+ "epoch": 24.33,
563
+ "learning_rate": 5.874269005847952e-05,
564
+ "loss": 0.3836,
565
+ "step": 73
566
+ },
567
+ {
568
+ "epoch": 24.67,
569
+ "learning_rate": 5.853801169590643e-05,
570
+ "loss": 0.3896,
571
+ "step": 74
572
+ },
573
+ {
574
+ "epoch": 25.0,
575
+ "learning_rate": 5.833333333333333e-05,
576
+ "loss": 0.5942,
577
+ "step": 75
578
+ },
579
+ {
580
+ "epoch": 25.33,
581
+ "learning_rate": 5.812865497076023e-05,
582
+ "loss": 0.3688,
583
+ "step": 76
584
+ },
585
+ {
586
+ "epoch": 25.67,
587
+ "learning_rate": 5.792397660818713e-05,
588
+ "loss": 0.3721,
589
+ "step": 77
590
+ },
591
+ {
592
+ "epoch": 26.0,
593
+ "learning_rate": 5.771929824561403e-05,
594
+ "loss": 0.5951,
595
+ "step": 78
596
+ },
597
+ {
598
+ "epoch": 26.33,
599
+ "learning_rate": 5.751461988304093e-05,
600
+ "loss": 0.3973,
601
+ "step": 79
602
+ },
603
+ {
604
+ "epoch": 26.67,
605
+ "learning_rate": 5.730994152046783e-05,
606
+ "loss": 0.3579,
607
+ "step": 80
608
+ },
609
+ {
610
+ "epoch": 26.67,
611
+ "eval_accuracy_dropoff": 0.40901049923036276,
612
+ "eval_accuracy_undropoff": 0.9908359437282933,
613
+ "eval_accuracy_unlabeled": NaN,
614
+ "eval_iou_dropoff": 0.34181486548107615,
615
+ "eval_iou_undropoff": 0.9644297203691943,
616
+ "eval_iou_unlabeled": 0.0,
617
+ "eval_loss": 0.3731442093849182,
618
+ "eval_mean_accuracy": 0.6999232214793281,
619
+ "eval_mean_iou": 0.43541486195009016,
620
+ "eval_overall_accuracy": 0.9650739034016927,
621
+ "eval_runtime": 1.7608,
622
+ "eval_samples_per_second": 8.519,
623
+ "eval_steps_per_second": 0.568,
624
+ "step": 80
625
+ }
626
+ ],
627
+ "max_steps": 360,
628
+ "num_train_epochs": 120,
629
+ "total_flos": 1.605671807090688e+16,
630
+ "trial_name": null,
631
+ "trial_params": null
632
+ }
checkpoint-80/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:824636194a0701719439cdb7771a1a1a2d7a8044b5c043163d7087355842dc71
3
+ size 4155
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dda72e49765313cff73b7df8836f48975c33abaaced3f0193b83810fbbcfb1cc
3
  size 14932813
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf3c492380c1369ce8f366fb3614bd6658de30c88484413eee29dd205e35e466
3
  size 14932813
runs/Feb12_06-10-28_robovision.csres.utexas.edu/events.out.tfevents.1707739842.robovision.csres.utexas.edu.2691839.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b39b9ed670ee6d85198a2f7f6be42f099b3a7b18527feca350c39c4b39e00b60
3
- size 14052
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34ab900d590a3d976d267ec7fe943f56f70a8b7ad3945031bc857eb81e9115cd
3
+ size 23328