FolashadeO commited on
Commit
91a09bc
·
verified ·
1 Parent(s): 47a1882

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. README.md +28 -0
  2. checkpoint-100/model.safetensors +3 -0
  3. checkpoint-100/optimizer.pt +3 -0
  4. checkpoint-100/rng_state.pth +3 -0
  5. checkpoint-100/scheduler.pt +3 -0
  6. checkpoint-100/trainer_state.json +72 -0
  7. checkpoint-100/training_args.bin +3 -0
  8. checkpoint-1000/model.safetensors +3 -0
  9. checkpoint-1000/optimizer.pt +3 -0
  10. checkpoint-1000/rng_state.pth +3 -0
  11. checkpoint-1000/scheduler.pt +3 -0
  12. checkpoint-1000/trainer_state.json +414 -0
  13. checkpoint-1000/training_args.bin +3 -0
  14. checkpoint-200/model.safetensors +3 -0
  15. checkpoint-200/optimizer.pt +3 -0
  16. checkpoint-200/rng_state.pth +3 -0
  17. checkpoint-200/scheduler.pt +3 -0
  18. checkpoint-200/trainer_state.json +110 -0
  19. checkpoint-200/training_args.bin +3 -0
  20. checkpoint-300/model.safetensors +3 -0
  21. checkpoint-300/optimizer.pt +3 -0
  22. checkpoint-300/rng_state.pth +3 -0
  23. checkpoint-300/scheduler.pt +3 -0
  24. checkpoint-300/trainer_state.json +148 -0
  25. checkpoint-300/training_args.bin +3 -0
  26. checkpoint-400/model.safetensors +3 -0
  27. checkpoint-400/optimizer.pt +3 -0
  28. checkpoint-400/rng_state.pth +3 -0
  29. checkpoint-400/scheduler.pt +3 -0
  30. checkpoint-400/trainer_state.json +186 -0
  31. checkpoint-400/training_args.bin +3 -0
  32. checkpoint-500/model.safetensors +3 -0
  33. checkpoint-500/optimizer.pt +3 -0
  34. checkpoint-500/rng_state.pth +3 -0
  35. checkpoint-500/scheduler.pt +3 -0
  36. checkpoint-500/trainer_state.json +224 -0
  37. checkpoint-500/training_args.bin +3 -0
  38. checkpoint-600/model.safetensors +3 -0
  39. checkpoint-600/optimizer.pt +3 -0
  40. checkpoint-600/rng_state.pth +3 -0
  41. checkpoint-600/scheduler.pt +3 -0
  42. checkpoint-600/trainer_state.json +262 -0
  43. checkpoint-600/training_args.bin +3 -0
  44. checkpoint-700/model.safetensors +3 -0
  45. checkpoint-700/optimizer.pt +3 -0
  46. checkpoint-700/rng_state.pth +3 -0
  47. checkpoint-700/scheduler.pt +3 -0
  48. checkpoint-700/trainer_state.json +300 -0
  49. checkpoint-700/training_args.bin +3 -0
  50. checkpoint-800/model.safetensors +3 -0
README.md ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ license: mit
4
+ tags:
5
+ - image-segmentation
6
+ - medical-imaging
7
+ - pytorch
8
+ - unet
9
+ datasets:
10
+ - Angelou0516/kvasir-seg
11
+ ---
12
+
13
+ # FolashadeO/week4
14
+
15
+ This model is a compact PyTorch U-Net for binary gastrointestinal polyp
16
+ segmentation on the Kvasir-SEG dataset.
17
+
18
+ ## Architecture
19
+
20
+ The network uses a four-level U-Net with convolutional encoder blocks,
21
+ a bottleneck, transposed-convolution decoder blocks, skip connections,
22
+ and a final `1 x 1` convolution that produces one mask logit channel.
23
+
24
+ ## Loss
25
+
26
+ Training uses a combined Dice and binary cross-entropy loss. BCE handles
27
+ per-pixel classification, while Dice loss directly optimizes mask overlap
28
+ and helps with foreground/background class imbalance.
checkpoint-100/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ac50980ebc8eb6587ce3dbd8fddcf798a0755a82d3fe758e0d4d734ec760c02
3
+ size 124209532
checkpoint-100/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:992d00dfd2985ef4ec37e30dba5621dd9d983fe61642987141e98292297f9011
3
+ size 248356555
checkpoint-100/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:098b29492211804ab324a36f37466821d948280bb74fce4ba895c03f13ecd878
3
+ size 14645
checkpoint-100/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:027203cd324846f900163fad9eecbe4aeb19c35a55cd188f85a03ff3e0f62d81
3
+ size 1465
checkpoint-100/trainer_state.json ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 100,
3
+ "best_metric": 0.4649257596145264,
4
+ "best_model_checkpoint": "./kvasir-unet/checkpoint-100",
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 100,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.25,
14
+ "grad_norm": 1.4807058572769165,
15
+ "learning_rate": 9.76e-05,
16
+ "loss": 1.2226847076416016,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.5,
21
+ "grad_norm": 1.6061848402023315,
22
+ "learning_rate": 9.51e-05,
23
+ "loss": 1.1226789093017577,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.75,
28
+ "grad_norm": 1.5647845268249512,
29
+ "learning_rate": 9.260000000000001e-05,
30
+ "loss": 1.0822339630126954,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 1.0,
35
+ "grad_norm": 4.204342365264893,
36
+ "learning_rate": 9.010000000000001e-05,
37
+ "loss": 1.00395751953125,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 1.0,
42
+ "eval_dice": 0.4649257596145264,
43
+ "eval_iou": 0.3028688409242743,
44
+ "eval_loss": 1.0397001504898071,
45
+ "eval_runtime": 4.527,
46
+ "eval_samples_per_second": 22.09,
47
+ "eval_steps_per_second": 2.872,
48
+ "step": 100
49
+ }
50
+ ],
51
+ "logging_steps": 25,
52
+ "max_steps": 1000,
53
+ "num_input_tokens_seen": 0,
54
+ "num_train_epochs": 10,
55
+ "save_steps": 500,
56
+ "stateful_callbacks": {
57
+ "TrainerControl": {
58
+ "args": {
59
+ "should_epoch_stop": false,
60
+ "should_evaluate": false,
61
+ "should_log": false,
62
+ "should_save": true,
63
+ "should_training_stop": false
64
+ },
65
+ "attributes": {}
66
+ }
67
+ },
68
+ "total_flos": 0.0,
69
+ "train_batch_size": 8,
70
+ "trial_name": null,
71
+ "trial_params": null
72
+ }
checkpoint-100/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04792f1700f54edbe0f85471cd1c5824c640ef6ced03e54f94ffcf8b7d12faea
3
+ size 5201
checkpoint-1000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:959ba1e85079688a2d7023bf6d0bd785f691070058cf5559a165aa40f2ffc21f
3
+ size 124209532
checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e3875e8db516e5bb464be6740318a33c38c14c1e3a1068da4c4fe1512d9a53d
3
+ size 248356555
checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed3753ab7977739b8eda494dd72defae5750f7283141b11a8f562160ba4c1a23
3
+ size 14645
checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9008af82ea2c682c9b64de9d5bd271649247b8bd3c3761983a4def45d64125f2
3
+ size 1465
checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,414 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1000,
3
+ "best_metric": 0.7648203854924553,
4
+ "best_model_checkpoint": "./kvasir-unet/checkpoint-1000",
5
+ "epoch": 10.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.25,
14
+ "grad_norm": 1.4807058572769165,
15
+ "learning_rate": 9.76e-05,
16
+ "loss": 1.2226847076416016,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.5,
21
+ "grad_norm": 1.6061848402023315,
22
+ "learning_rate": 9.51e-05,
23
+ "loss": 1.1226789093017577,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.75,
28
+ "grad_norm": 1.5647845268249512,
29
+ "learning_rate": 9.260000000000001e-05,
30
+ "loss": 1.0822339630126954,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 1.0,
35
+ "grad_norm": 4.204342365264893,
36
+ "learning_rate": 9.010000000000001e-05,
37
+ "loss": 1.00395751953125,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 1.0,
42
+ "eval_dice": 0.4649257596145264,
43
+ "eval_iou": 0.3028688409242743,
44
+ "eval_loss": 1.0397001504898071,
45
+ "eval_runtime": 4.527,
46
+ "eval_samples_per_second": 22.09,
47
+ "eval_steps_per_second": 2.872,
48
+ "step": 100
49
+ },
50
+ {
51
+ "epoch": 1.25,
52
+ "grad_norm": 2.125596284866333,
53
+ "learning_rate": 8.76e-05,
54
+ "loss": 0.9969179534912109,
55
+ "step": 125
56
+ },
57
+ {
58
+ "epoch": 1.5,
59
+ "grad_norm": 2.8102705478668213,
60
+ "learning_rate": 8.510000000000001e-05,
61
+ "loss": 0.9738675689697266,
62
+ "step": 150
63
+ },
64
+ {
65
+ "epoch": 1.75,
66
+ "grad_norm": 2.3319432735443115,
67
+ "learning_rate": 8.26e-05,
68
+ "loss": 0.9797834777832031,
69
+ "step": 175
70
+ },
71
+ {
72
+ "epoch": 2.0,
73
+ "grad_norm": 3.546826124191284,
74
+ "learning_rate": 8.010000000000001e-05,
75
+ "loss": 0.936845932006836,
76
+ "step": 200
77
+ },
78
+ {
79
+ "epoch": 2.0,
80
+ "eval_dice": 0.5913469032255761,
81
+ "eval_iou": 0.41979614051111847,
82
+ "eval_loss": 0.9874243140220642,
83
+ "eval_runtime": 4.4621,
84
+ "eval_samples_per_second": 22.411,
85
+ "eval_steps_per_second": 2.913,
86
+ "step": 200
87
+ },
88
+ {
89
+ "epoch": 2.25,
90
+ "grad_norm": 4.708238124847412,
91
+ "learning_rate": 7.76e-05,
92
+ "loss": 0.9107416534423828,
93
+ "step": 225
94
+ },
95
+ {
96
+ "epoch": 2.5,
97
+ "grad_norm": 1.9260565042495728,
98
+ "learning_rate": 7.510000000000001e-05,
99
+ "loss": 0.9003370666503906,
100
+ "step": 250
101
+ },
102
+ {
103
+ "epoch": 2.75,
104
+ "grad_norm": 2.7347285747528076,
105
+ "learning_rate": 7.26e-05,
106
+ "loss": 0.8682646942138672,
107
+ "step": 275
108
+ },
109
+ {
110
+ "epoch": 3.0,
111
+ "grad_norm": 1.9796888828277588,
112
+ "learning_rate": 7.01e-05,
113
+ "loss": 0.8491334533691406,
114
+ "step": 300
115
+ },
116
+ {
117
+ "epoch": 3.0,
118
+ "eval_dice": 0.6287870060272235,
119
+ "eval_iou": 0.45856278594402466,
120
+ "eval_loss": 0.880113422870636,
121
+ "eval_runtime": 4.6294,
122
+ "eval_samples_per_second": 21.601,
123
+ "eval_steps_per_second": 2.808,
124
+ "step": 300
125
+ },
126
+ {
127
+ "epoch": 3.25,
128
+ "grad_norm": 4.970420837402344,
129
+ "learning_rate": 6.76e-05,
130
+ "loss": 0.8522591400146484,
131
+ "step": 325
132
+ },
133
+ {
134
+ "epoch": 3.5,
135
+ "grad_norm": 2.374986171722412,
136
+ "learning_rate": 6.510000000000001e-05,
137
+ "loss": 0.7948677825927735,
138
+ "step": 350
139
+ },
140
+ {
141
+ "epoch": 3.75,
142
+ "grad_norm": 5.10560417175293,
143
+ "learning_rate": 6.26e-05,
144
+ "loss": 0.8352813720703125,
145
+ "step": 375
146
+ },
147
+ {
148
+ "epoch": 4.0,
149
+ "grad_norm": 4.5725274085998535,
150
+ "learning_rate": 6.0100000000000004e-05,
151
+ "loss": 0.7697490692138672,
152
+ "step": 400
153
+ },
154
+ {
155
+ "epoch": 4.0,
156
+ "eval_dice": 0.6448364361669014,
157
+ "eval_iou": 0.4758367704940324,
158
+ "eval_loss": 0.8459730744361877,
159
+ "eval_runtime": 4.4418,
160
+ "eval_samples_per_second": 22.514,
161
+ "eval_steps_per_second": 2.927,
162
+ "step": 400
163
+ },
164
+ {
165
+ "epoch": 4.25,
166
+ "grad_norm": 7.802286148071289,
167
+ "learning_rate": 5.76e-05,
168
+ "loss": 0.7570986175537109,
169
+ "step": 425
170
+ },
171
+ {
172
+ "epoch": 4.5,
173
+ "grad_norm": 7.580319881439209,
174
+ "learning_rate": 5.5100000000000004e-05,
175
+ "loss": 0.763985366821289,
176
+ "step": 450
177
+ },
178
+ {
179
+ "epoch": 4.75,
180
+ "grad_norm": 3.051490068435669,
181
+ "learning_rate": 5.2600000000000005e-05,
182
+ "loss": 0.7334027862548829,
183
+ "step": 475
184
+ },
185
+ {
186
+ "epoch": 5.0,
187
+ "grad_norm": 3.4101736545562744,
188
+ "learning_rate": 5.0100000000000005e-05,
189
+ "loss": 0.7454586791992187,
190
+ "step": 500
191
+ },
192
+ {
193
+ "epoch": 5.0,
194
+ "eval_dice": 0.6826710522273053,
195
+ "eval_iou": 0.51822387496614,
196
+ "eval_loss": 0.7638939619064331,
197
+ "eval_runtime": 4.4461,
198
+ "eval_samples_per_second": 22.491,
199
+ "eval_steps_per_second": 2.924,
200
+ "step": 500
201
+ },
202
+ {
203
+ "epoch": 5.25,
204
+ "grad_norm": 9.261576652526855,
205
+ "learning_rate": 4.76e-05,
206
+ "loss": 0.7240135955810547,
207
+ "step": 525
208
+ },
209
+ {
210
+ "epoch": 5.5,
211
+ "grad_norm": 3.081444501876831,
212
+ "learning_rate": 4.5100000000000005e-05,
213
+ "loss": 0.7151520538330078,
214
+ "step": 550
215
+ },
216
+ {
217
+ "epoch": 5.75,
218
+ "grad_norm": 4.129289150238037,
219
+ "learning_rate": 4.26e-05,
220
+ "loss": 0.6827881622314453,
221
+ "step": 575
222
+ },
223
+ {
224
+ "epoch": 6.0,
225
+ "grad_norm": 3.271570920944214,
226
+ "learning_rate": 4.0100000000000006e-05,
227
+ "loss": 0.6640605926513672,
228
+ "step": 600
229
+ },
230
+ {
231
+ "epoch": 6.0,
232
+ "eval_dice": 0.7196315708151912,
233
+ "eval_iou": 0.562050553478283,
234
+ "eval_loss": 0.7080976366996765,
235
+ "eval_runtime": 4.5035,
236
+ "eval_samples_per_second": 22.205,
237
+ "eval_steps_per_second": 2.887,
238
+ "step": 600
239
+ },
240
+ {
241
+ "epoch": 6.25,
242
+ "grad_norm": 3.007444143295288,
243
+ "learning_rate": 3.76e-05,
244
+ "loss": 0.6774704742431641,
245
+ "step": 625
246
+ },
247
+ {
248
+ "epoch": 6.5,
249
+ "grad_norm": 6.9112420082092285,
250
+ "learning_rate": 3.51e-05,
251
+ "loss": 0.6406173706054688,
252
+ "step": 650
253
+ },
254
+ {
255
+ "epoch": 6.75,
256
+ "grad_norm": 3.0203912258148193,
257
+ "learning_rate": 3.26e-05,
258
+ "loss": 0.6287059783935547,
259
+ "step": 675
260
+ },
261
+ {
262
+ "epoch": 7.0,
263
+ "grad_norm": 3.216691732406616,
264
+ "learning_rate": 3.01e-05,
265
+ "loss": 0.647440185546875,
266
+ "step": 700
267
+ },
268
+ {
269
+ "epoch": 7.0,
270
+ "eval_dice": 0.7246417440175368,
271
+ "eval_iou": 0.5681869886775344,
272
+ "eval_loss": 0.691881000995636,
273
+ "eval_runtime": 4.5083,
274
+ "eval_samples_per_second": 22.181,
275
+ "eval_steps_per_second": 2.884,
276
+ "step": 700
277
+ },
278
+ {
279
+ "epoch": 7.25,
280
+ "grad_norm": 5.566691875457764,
281
+ "learning_rate": 2.7600000000000003e-05,
282
+ "loss": 0.640998306274414,
283
+ "step": 725
284
+ },
285
+ {
286
+ "epoch": 7.5,
287
+ "grad_norm": 1.7265862226486206,
288
+ "learning_rate": 2.51e-05,
289
+ "loss": 0.6094163513183594,
290
+ "step": 750
291
+ },
292
+ {
293
+ "epoch": 7.75,
294
+ "grad_norm": 4.541012763977051,
295
+ "learning_rate": 2.26e-05,
296
+ "loss": 0.6245226287841796,
297
+ "step": 775
298
+ },
299
+ {
300
+ "epoch": 8.0,
301
+ "grad_norm": 3.6882596015930176,
302
+ "learning_rate": 2.01e-05,
303
+ "loss": 0.56927001953125,
304
+ "step": 800
305
+ },
306
+ {
307
+ "epoch": 8.0,
308
+ "eval_dice": 0.7364048604227745,
309
+ "eval_iou": 0.5827856171617375,
310
+ "eval_loss": 0.6652660965919495,
311
+ "eval_runtime": 4.5763,
312
+ "eval_samples_per_second": 21.852,
313
+ "eval_steps_per_second": 2.841,
314
+ "step": 800
315
+ },
316
+ {
317
+ "epoch": 8.25,
318
+ "grad_norm": 5.979460716247559,
319
+ "learning_rate": 1.76e-05,
320
+ "loss": 0.5805312347412109,
321
+ "step": 825
322
+ },
323
+ {
324
+ "epoch": 8.5,
325
+ "grad_norm": 5.097400188446045,
326
+ "learning_rate": 1.51e-05,
327
+ "loss": 0.5637349700927734,
328
+ "step": 850
329
+ },
330
+ {
331
+ "epoch": 8.75,
332
+ "grad_norm": 3.3682267665863037,
333
+ "learning_rate": 1.2600000000000001e-05,
334
+ "loss": 0.5543782806396484,
335
+ "step": 875
336
+ },
337
+ {
338
+ "epoch": 9.0,
339
+ "grad_norm": 4.063240051269531,
340
+ "learning_rate": 1.0100000000000002e-05,
341
+ "loss": 0.5821345520019531,
342
+ "step": 900
343
+ },
344
+ {
345
+ "epoch": 9.0,
346
+ "eval_dice": 0.7544327435690541,
347
+ "eval_iou": 0.6056942714502439,
348
+ "eval_loss": 0.6264443397521973,
349
+ "eval_runtime": 4.4354,
350
+ "eval_samples_per_second": 22.546,
351
+ "eval_steps_per_second": 2.931,
352
+ "step": 900
353
+ },
354
+ {
355
+ "epoch": 9.25,
356
+ "grad_norm": 3.4463164806365967,
357
+ "learning_rate": 7.6e-06,
358
+ "loss": 0.5670050048828125,
359
+ "step": 925
360
+ },
361
+ {
362
+ "epoch": 9.5,
363
+ "grad_norm": 2.3057284355163574,
364
+ "learning_rate": 5.1e-06,
365
+ "loss": 0.5416771697998047,
366
+ "step": 950
367
+ },
368
+ {
369
+ "epoch": 9.75,
370
+ "grad_norm": 6.042228698730469,
371
+ "learning_rate": 2.6e-06,
372
+ "loss": 0.5341218566894531,
373
+ "step": 975
374
+ },
375
+ {
376
+ "epoch": 10.0,
377
+ "grad_norm": 1.7755556106567383,
378
+ "learning_rate": 1.0000000000000001e-07,
379
+ "loss": 0.5268088150024414,
380
+ "step": 1000
381
+ },
382
+ {
383
+ "epoch": 10.0,
384
+ "eval_dice": 0.7648203854924553,
385
+ "eval_iou": 0.6191978699585076,
386
+ "eval_loss": 0.6198652386665344,
387
+ "eval_runtime": 4.5227,
388
+ "eval_samples_per_second": 22.111,
389
+ "eval_steps_per_second": 2.874,
390
+ "step": 1000
391
+ }
392
+ ],
393
+ "logging_steps": 25,
394
+ "max_steps": 1000,
395
+ "num_input_tokens_seen": 0,
396
+ "num_train_epochs": 10,
397
+ "save_steps": 500,
398
+ "stateful_callbacks": {
399
+ "TrainerControl": {
400
+ "args": {
401
+ "should_epoch_stop": false,
402
+ "should_evaluate": false,
403
+ "should_log": false,
404
+ "should_save": true,
405
+ "should_training_stop": true
406
+ },
407
+ "attributes": {}
408
+ }
409
+ },
410
+ "total_flos": 0.0,
411
+ "train_batch_size": 8,
412
+ "trial_name": null,
413
+ "trial_params": null
414
+ }
checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04792f1700f54edbe0f85471cd1c5824c640ef6ced03e54f94ffcf8b7d12faea
3
+ size 5201
checkpoint-200/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44426de5d0279ae93f64c0ff08650a2dab79f3dac8a23b277c9e683ae6a05634
3
+ size 124209532
checkpoint-200/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d265054cb900b7094002aeb92899b9838dccb7db97641b66e2ac6a7bbd6f83b
3
+ size 248356555
checkpoint-200/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8e2011629d8bed3ef560fa11175cac55684c4e12a72634bb24abf767b6c7399
3
+ size 14645
checkpoint-200/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:793b45e356907c0dd0ce0a30dee131ac2edb92e56cb56cf8c95af4feabe3d5a8
3
+ size 1465
checkpoint-200/trainer_state.json ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 200,
3
+ "best_metric": 0.5913469032255761,
4
+ "best_model_checkpoint": "./kvasir-unet/checkpoint-200",
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 200,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.25,
14
+ "grad_norm": 1.4807058572769165,
15
+ "learning_rate": 9.76e-05,
16
+ "loss": 1.2226847076416016,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.5,
21
+ "grad_norm": 1.6061848402023315,
22
+ "learning_rate": 9.51e-05,
23
+ "loss": 1.1226789093017577,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.75,
28
+ "grad_norm": 1.5647845268249512,
29
+ "learning_rate": 9.260000000000001e-05,
30
+ "loss": 1.0822339630126954,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 1.0,
35
+ "grad_norm": 4.204342365264893,
36
+ "learning_rate": 9.010000000000001e-05,
37
+ "loss": 1.00395751953125,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 1.0,
42
+ "eval_dice": 0.4649257596145264,
43
+ "eval_iou": 0.3028688409242743,
44
+ "eval_loss": 1.0397001504898071,
45
+ "eval_runtime": 4.527,
46
+ "eval_samples_per_second": 22.09,
47
+ "eval_steps_per_second": 2.872,
48
+ "step": 100
49
+ },
50
+ {
51
+ "epoch": 1.25,
52
+ "grad_norm": 2.125596284866333,
53
+ "learning_rate": 8.76e-05,
54
+ "loss": 0.9969179534912109,
55
+ "step": 125
56
+ },
57
+ {
58
+ "epoch": 1.5,
59
+ "grad_norm": 2.8102705478668213,
60
+ "learning_rate": 8.510000000000001e-05,
61
+ "loss": 0.9738675689697266,
62
+ "step": 150
63
+ },
64
+ {
65
+ "epoch": 1.75,
66
+ "grad_norm": 2.3319432735443115,
67
+ "learning_rate": 8.26e-05,
68
+ "loss": 0.9797834777832031,
69
+ "step": 175
70
+ },
71
+ {
72
+ "epoch": 2.0,
73
+ "grad_norm": 3.546826124191284,
74
+ "learning_rate": 8.010000000000001e-05,
75
+ "loss": 0.936845932006836,
76
+ "step": 200
77
+ },
78
+ {
79
+ "epoch": 2.0,
80
+ "eval_dice": 0.5913469032255761,
81
+ "eval_iou": 0.41979614051111847,
82
+ "eval_loss": 0.9874243140220642,
83
+ "eval_runtime": 4.4621,
84
+ "eval_samples_per_second": 22.411,
85
+ "eval_steps_per_second": 2.913,
86
+ "step": 200
87
+ }
88
+ ],
89
+ "logging_steps": 25,
90
+ "max_steps": 1000,
91
+ "num_input_tokens_seen": 0,
92
+ "num_train_epochs": 10,
93
+ "save_steps": 500,
94
+ "stateful_callbacks": {
95
+ "TrainerControl": {
96
+ "args": {
97
+ "should_epoch_stop": false,
98
+ "should_evaluate": false,
99
+ "should_log": false,
100
+ "should_save": true,
101
+ "should_training_stop": false
102
+ },
103
+ "attributes": {}
104
+ }
105
+ },
106
+ "total_flos": 0.0,
107
+ "train_batch_size": 8,
108
+ "trial_name": null,
109
+ "trial_params": null
110
+ }
checkpoint-200/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04792f1700f54edbe0f85471cd1c5824c640ef6ced03e54f94ffcf8b7d12faea
3
+ size 5201
checkpoint-300/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0115699bedfbb268a96b914377542b23413a5977d1c25f052c55fc9f6c66737f
3
+ size 124209532
checkpoint-300/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:649c9d2775a7d0d739e3710b6d568cc6e32b656552e94d818f51b6b934aa518f
3
+ size 248356555
checkpoint-300/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01f9a0f7843a37be87edd23f4e88aa93b38b95cc2c07503eeb1cf2e4632453a2
3
+ size 14645
checkpoint-300/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0fd1199dc154edf23304b20c74883b404bdec174b48bea5676d181b512de88bb
3
+ size 1465
checkpoint-300/trainer_state.json ADDED
@@ -0,0 +1,148 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 300,
3
+ "best_metric": 0.6287870060272235,
4
+ "best_model_checkpoint": "./kvasir-unet/checkpoint-300",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 300,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.25,
14
+ "grad_norm": 1.4807058572769165,
15
+ "learning_rate": 9.76e-05,
16
+ "loss": 1.2226847076416016,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.5,
21
+ "grad_norm": 1.6061848402023315,
22
+ "learning_rate": 9.51e-05,
23
+ "loss": 1.1226789093017577,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.75,
28
+ "grad_norm": 1.5647845268249512,
29
+ "learning_rate": 9.260000000000001e-05,
30
+ "loss": 1.0822339630126954,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 1.0,
35
+ "grad_norm": 4.204342365264893,
36
+ "learning_rate": 9.010000000000001e-05,
37
+ "loss": 1.00395751953125,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 1.0,
42
+ "eval_dice": 0.4649257596145264,
43
+ "eval_iou": 0.3028688409242743,
44
+ "eval_loss": 1.0397001504898071,
45
+ "eval_runtime": 4.527,
46
+ "eval_samples_per_second": 22.09,
47
+ "eval_steps_per_second": 2.872,
48
+ "step": 100
49
+ },
50
+ {
51
+ "epoch": 1.25,
52
+ "grad_norm": 2.125596284866333,
53
+ "learning_rate": 8.76e-05,
54
+ "loss": 0.9969179534912109,
55
+ "step": 125
56
+ },
57
+ {
58
+ "epoch": 1.5,
59
+ "grad_norm": 2.8102705478668213,
60
+ "learning_rate": 8.510000000000001e-05,
61
+ "loss": 0.9738675689697266,
62
+ "step": 150
63
+ },
64
+ {
65
+ "epoch": 1.75,
66
+ "grad_norm": 2.3319432735443115,
67
+ "learning_rate": 8.26e-05,
68
+ "loss": 0.9797834777832031,
69
+ "step": 175
70
+ },
71
+ {
72
+ "epoch": 2.0,
73
+ "grad_norm": 3.546826124191284,
74
+ "learning_rate": 8.010000000000001e-05,
75
+ "loss": 0.936845932006836,
76
+ "step": 200
77
+ },
78
+ {
79
+ "epoch": 2.0,
80
+ "eval_dice": 0.5913469032255761,
81
+ "eval_iou": 0.41979614051111847,
82
+ "eval_loss": 0.9874243140220642,
83
+ "eval_runtime": 4.4621,
84
+ "eval_samples_per_second": 22.411,
85
+ "eval_steps_per_second": 2.913,
86
+ "step": 200
87
+ },
88
+ {
89
+ "epoch": 2.25,
90
+ "grad_norm": 4.708238124847412,
91
+ "learning_rate": 7.76e-05,
92
+ "loss": 0.9107416534423828,
93
+ "step": 225
94
+ },
95
+ {
96
+ "epoch": 2.5,
97
+ "grad_norm": 1.9260565042495728,
98
+ "learning_rate": 7.510000000000001e-05,
99
+ "loss": 0.9003370666503906,
100
+ "step": 250
101
+ },
102
+ {
103
+ "epoch": 2.75,
104
+ "grad_norm": 2.7347285747528076,
105
+ "learning_rate": 7.26e-05,
106
+ "loss": 0.8682646942138672,
107
+ "step": 275
108
+ },
109
+ {
110
+ "epoch": 3.0,
111
+ "grad_norm": 1.9796888828277588,
112
+ "learning_rate": 7.01e-05,
113
+ "loss": 0.8491334533691406,
114
+ "step": 300
115
+ },
116
+ {
117
+ "epoch": 3.0,
118
+ "eval_dice": 0.6287870060272235,
119
+ "eval_iou": 0.45856278594402466,
120
+ "eval_loss": 0.880113422870636,
121
+ "eval_runtime": 4.6294,
122
+ "eval_samples_per_second": 21.601,
123
+ "eval_steps_per_second": 2.808,
124
+ "step": 300
125
+ }
126
+ ],
127
+ "logging_steps": 25,
128
+ "max_steps": 1000,
129
+ "num_input_tokens_seen": 0,
130
+ "num_train_epochs": 10,
131
+ "save_steps": 500,
132
+ "stateful_callbacks": {
133
+ "TrainerControl": {
134
+ "args": {
135
+ "should_epoch_stop": false,
136
+ "should_evaluate": false,
137
+ "should_log": false,
138
+ "should_save": true,
139
+ "should_training_stop": false
140
+ },
141
+ "attributes": {}
142
+ }
143
+ },
144
+ "total_flos": 0.0,
145
+ "train_batch_size": 8,
146
+ "trial_name": null,
147
+ "trial_params": null
148
+ }
checkpoint-300/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04792f1700f54edbe0f85471cd1c5824c640ef6ced03e54f94ffcf8b7d12faea
3
+ size 5201
checkpoint-400/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a92a732c4926ab86d715918a1712320d18a352657b6891b8ea8d9a3ac6026067
3
+ size 124209532
checkpoint-400/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:49de9e8c7f785dfde1ed38ab3878314d7b18a1a7126750df7831715888c7a0e6
3
+ size 248356555
checkpoint-400/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fdc37bbd2e979f041dfbbb004a5c74bab6cdda159cb18116df728588515a9ef6
3
+ size 14645
checkpoint-400/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69df656554bba9e77df00d6792762dde1865a2df0e7d1dc6df12c78c21791c28
3
+ size 1465
checkpoint-400/trainer_state.json ADDED
@@ -0,0 +1,186 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 400,
3
+ "best_metric": 0.6448364361669014,
4
+ "best_model_checkpoint": "./kvasir-unet/checkpoint-400",
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 400,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.25,
14
+ "grad_norm": 1.4807058572769165,
15
+ "learning_rate": 9.76e-05,
16
+ "loss": 1.2226847076416016,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.5,
21
+ "grad_norm": 1.6061848402023315,
22
+ "learning_rate": 9.51e-05,
23
+ "loss": 1.1226789093017577,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.75,
28
+ "grad_norm": 1.5647845268249512,
29
+ "learning_rate": 9.260000000000001e-05,
30
+ "loss": 1.0822339630126954,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 1.0,
35
+ "grad_norm": 4.204342365264893,
36
+ "learning_rate": 9.010000000000001e-05,
37
+ "loss": 1.00395751953125,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 1.0,
42
+ "eval_dice": 0.4649257596145264,
43
+ "eval_iou": 0.3028688409242743,
44
+ "eval_loss": 1.0397001504898071,
45
+ "eval_runtime": 4.527,
46
+ "eval_samples_per_second": 22.09,
47
+ "eval_steps_per_second": 2.872,
48
+ "step": 100
49
+ },
50
+ {
51
+ "epoch": 1.25,
52
+ "grad_norm": 2.125596284866333,
53
+ "learning_rate": 8.76e-05,
54
+ "loss": 0.9969179534912109,
55
+ "step": 125
56
+ },
57
+ {
58
+ "epoch": 1.5,
59
+ "grad_norm": 2.8102705478668213,
60
+ "learning_rate": 8.510000000000001e-05,
61
+ "loss": 0.9738675689697266,
62
+ "step": 150
63
+ },
64
+ {
65
+ "epoch": 1.75,
66
+ "grad_norm": 2.3319432735443115,
67
+ "learning_rate": 8.26e-05,
68
+ "loss": 0.9797834777832031,
69
+ "step": 175
70
+ },
71
+ {
72
+ "epoch": 2.0,
73
+ "grad_norm": 3.546826124191284,
74
+ "learning_rate": 8.010000000000001e-05,
75
+ "loss": 0.936845932006836,
76
+ "step": 200
77
+ },
78
+ {
79
+ "epoch": 2.0,
80
+ "eval_dice": 0.5913469032255761,
81
+ "eval_iou": 0.41979614051111847,
82
+ "eval_loss": 0.9874243140220642,
83
+ "eval_runtime": 4.4621,
84
+ "eval_samples_per_second": 22.411,
85
+ "eval_steps_per_second": 2.913,
86
+ "step": 200
87
+ },
88
+ {
89
+ "epoch": 2.25,
90
+ "grad_norm": 4.708238124847412,
91
+ "learning_rate": 7.76e-05,
92
+ "loss": 0.9107416534423828,
93
+ "step": 225
94
+ },
95
+ {
96
+ "epoch": 2.5,
97
+ "grad_norm": 1.9260565042495728,
98
+ "learning_rate": 7.510000000000001e-05,
99
+ "loss": 0.9003370666503906,
100
+ "step": 250
101
+ },
102
+ {
103
+ "epoch": 2.75,
104
+ "grad_norm": 2.7347285747528076,
105
+ "learning_rate": 7.26e-05,
106
+ "loss": 0.8682646942138672,
107
+ "step": 275
108
+ },
109
+ {
110
+ "epoch": 3.0,
111
+ "grad_norm": 1.9796888828277588,
112
+ "learning_rate": 7.01e-05,
113
+ "loss": 0.8491334533691406,
114
+ "step": 300
115
+ },
116
+ {
117
+ "epoch": 3.0,
118
+ "eval_dice": 0.6287870060272235,
119
+ "eval_iou": 0.45856278594402466,
120
+ "eval_loss": 0.880113422870636,
121
+ "eval_runtime": 4.6294,
122
+ "eval_samples_per_second": 21.601,
123
+ "eval_steps_per_second": 2.808,
124
+ "step": 300
125
+ },
126
+ {
127
+ "epoch": 3.25,
128
+ "grad_norm": 4.970420837402344,
129
+ "learning_rate": 6.76e-05,
130
+ "loss": 0.8522591400146484,
131
+ "step": 325
132
+ },
133
+ {
134
+ "epoch": 3.5,
135
+ "grad_norm": 2.374986171722412,
136
+ "learning_rate": 6.510000000000001e-05,
137
+ "loss": 0.7948677825927735,
138
+ "step": 350
139
+ },
140
+ {
141
+ "epoch": 3.75,
142
+ "grad_norm": 5.10560417175293,
143
+ "learning_rate": 6.26e-05,
144
+ "loss": 0.8352813720703125,
145
+ "step": 375
146
+ },
147
+ {
148
+ "epoch": 4.0,
149
+ "grad_norm": 4.5725274085998535,
150
+ "learning_rate": 6.0100000000000004e-05,
151
+ "loss": 0.7697490692138672,
152
+ "step": 400
153
+ },
154
+ {
155
+ "epoch": 4.0,
156
+ "eval_dice": 0.6448364361669014,
157
+ "eval_iou": 0.4758367704940324,
158
+ "eval_loss": 0.8459730744361877,
159
+ "eval_runtime": 4.4418,
160
+ "eval_samples_per_second": 22.514,
161
+ "eval_steps_per_second": 2.927,
162
+ "step": 400
163
+ }
164
+ ],
165
+ "logging_steps": 25,
166
+ "max_steps": 1000,
167
+ "num_input_tokens_seen": 0,
168
+ "num_train_epochs": 10,
169
+ "save_steps": 500,
170
+ "stateful_callbacks": {
171
+ "TrainerControl": {
172
+ "args": {
173
+ "should_epoch_stop": false,
174
+ "should_evaluate": false,
175
+ "should_log": false,
176
+ "should_save": true,
177
+ "should_training_stop": false
178
+ },
179
+ "attributes": {}
180
+ }
181
+ },
182
+ "total_flos": 0.0,
183
+ "train_batch_size": 8,
184
+ "trial_name": null,
185
+ "trial_params": null
186
+ }
checkpoint-400/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04792f1700f54edbe0f85471cd1c5824c640ef6ced03e54f94ffcf8b7d12faea
3
+ size 5201
checkpoint-500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3ae0e946e16ae2aeb75d83e7bbc9701161b96f7de613caa3868a42501d6962b6
3
+ size 124209532
checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ffb2e2935b8a42ce6d95a401a505e2601c787fcc51a2cf1bc931f44227abc6ee
3
+ size 248356555
checkpoint-500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17cd930da9783ca70bad4b9cdeee6a06c0acea8f34645a333c93341f487f66a3
3
+ size 14645
checkpoint-500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:310b94cc33fed5e06d84d656e61170246da35f5377b1130f8d2df6d65c4047dc
3
+ size 1465
checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,224 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 500,
3
+ "best_metric": 0.6826710522273053,
4
+ "best_model_checkpoint": "./kvasir-unet/checkpoint-500",
5
+ "epoch": 5.0,
6
+ "eval_steps": 500,
7
+ "global_step": 500,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.25,
14
+ "grad_norm": 1.4807058572769165,
15
+ "learning_rate": 9.76e-05,
16
+ "loss": 1.2226847076416016,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.5,
21
+ "grad_norm": 1.6061848402023315,
22
+ "learning_rate": 9.51e-05,
23
+ "loss": 1.1226789093017577,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.75,
28
+ "grad_norm": 1.5647845268249512,
29
+ "learning_rate": 9.260000000000001e-05,
30
+ "loss": 1.0822339630126954,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 1.0,
35
+ "grad_norm": 4.204342365264893,
36
+ "learning_rate": 9.010000000000001e-05,
37
+ "loss": 1.00395751953125,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 1.0,
42
+ "eval_dice": 0.4649257596145264,
43
+ "eval_iou": 0.3028688409242743,
44
+ "eval_loss": 1.0397001504898071,
45
+ "eval_runtime": 4.527,
46
+ "eval_samples_per_second": 22.09,
47
+ "eval_steps_per_second": 2.872,
48
+ "step": 100
49
+ },
50
+ {
51
+ "epoch": 1.25,
52
+ "grad_norm": 2.125596284866333,
53
+ "learning_rate": 8.76e-05,
54
+ "loss": 0.9969179534912109,
55
+ "step": 125
56
+ },
57
+ {
58
+ "epoch": 1.5,
59
+ "grad_norm": 2.8102705478668213,
60
+ "learning_rate": 8.510000000000001e-05,
61
+ "loss": 0.9738675689697266,
62
+ "step": 150
63
+ },
64
+ {
65
+ "epoch": 1.75,
66
+ "grad_norm": 2.3319432735443115,
67
+ "learning_rate": 8.26e-05,
68
+ "loss": 0.9797834777832031,
69
+ "step": 175
70
+ },
71
+ {
72
+ "epoch": 2.0,
73
+ "grad_norm": 3.546826124191284,
74
+ "learning_rate": 8.010000000000001e-05,
75
+ "loss": 0.936845932006836,
76
+ "step": 200
77
+ },
78
+ {
79
+ "epoch": 2.0,
80
+ "eval_dice": 0.5913469032255761,
81
+ "eval_iou": 0.41979614051111847,
82
+ "eval_loss": 0.9874243140220642,
83
+ "eval_runtime": 4.4621,
84
+ "eval_samples_per_second": 22.411,
85
+ "eval_steps_per_second": 2.913,
86
+ "step": 200
87
+ },
88
+ {
89
+ "epoch": 2.25,
90
+ "grad_norm": 4.708238124847412,
91
+ "learning_rate": 7.76e-05,
92
+ "loss": 0.9107416534423828,
93
+ "step": 225
94
+ },
95
+ {
96
+ "epoch": 2.5,
97
+ "grad_norm": 1.9260565042495728,
98
+ "learning_rate": 7.510000000000001e-05,
99
+ "loss": 0.9003370666503906,
100
+ "step": 250
101
+ },
102
+ {
103
+ "epoch": 2.75,
104
+ "grad_norm": 2.7347285747528076,
105
+ "learning_rate": 7.26e-05,
106
+ "loss": 0.8682646942138672,
107
+ "step": 275
108
+ },
109
+ {
110
+ "epoch": 3.0,
111
+ "grad_norm": 1.9796888828277588,
112
+ "learning_rate": 7.01e-05,
113
+ "loss": 0.8491334533691406,
114
+ "step": 300
115
+ },
116
+ {
117
+ "epoch": 3.0,
118
+ "eval_dice": 0.6287870060272235,
119
+ "eval_iou": 0.45856278594402466,
120
+ "eval_loss": 0.880113422870636,
121
+ "eval_runtime": 4.6294,
122
+ "eval_samples_per_second": 21.601,
123
+ "eval_steps_per_second": 2.808,
124
+ "step": 300
125
+ },
126
+ {
127
+ "epoch": 3.25,
128
+ "grad_norm": 4.970420837402344,
129
+ "learning_rate": 6.76e-05,
130
+ "loss": 0.8522591400146484,
131
+ "step": 325
132
+ },
133
+ {
134
+ "epoch": 3.5,
135
+ "grad_norm": 2.374986171722412,
136
+ "learning_rate": 6.510000000000001e-05,
137
+ "loss": 0.7948677825927735,
138
+ "step": 350
139
+ },
140
+ {
141
+ "epoch": 3.75,
142
+ "grad_norm": 5.10560417175293,
143
+ "learning_rate": 6.26e-05,
144
+ "loss": 0.8352813720703125,
145
+ "step": 375
146
+ },
147
+ {
148
+ "epoch": 4.0,
149
+ "grad_norm": 4.5725274085998535,
150
+ "learning_rate": 6.0100000000000004e-05,
151
+ "loss": 0.7697490692138672,
152
+ "step": 400
153
+ },
154
+ {
155
+ "epoch": 4.0,
156
+ "eval_dice": 0.6448364361669014,
157
+ "eval_iou": 0.4758367704940324,
158
+ "eval_loss": 0.8459730744361877,
159
+ "eval_runtime": 4.4418,
160
+ "eval_samples_per_second": 22.514,
161
+ "eval_steps_per_second": 2.927,
162
+ "step": 400
163
+ },
164
+ {
165
+ "epoch": 4.25,
166
+ "grad_norm": 7.802286148071289,
167
+ "learning_rate": 5.76e-05,
168
+ "loss": 0.7570986175537109,
169
+ "step": 425
170
+ },
171
+ {
172
+ "epoch": 4.5,
173
+ "grad_norm": 7.580319881439209,
174
+ "learning_rate": 5.5100000000000004e-05,
175
+ "loss": 0.763985366821289,
176
+ "step": 450
177
+ },
178
+ {
179
+ "epoch": 4.75,
180
+ "grad_norm": 3.051490068435669,
181
+ "learning_rate": 5.2600000000000005e-05,
182
+ "loss": 0.7334027862548829,
183
+ "step": 475
184
+ },
185
+ {
186
+ "epoch": 5.0,
187
+ "grad_norm": 3.4101736545562744,
188
+ "learning_rate": 5.0100000000000005e-05,
189
+ "loss": 0.7454586791992187,
190
+ "step": 500
191
+ },
192
+ {
193
+ "epoch": 5.0,
194
+ "eval_dice": 0.6826710522273053,
195
+ "eval_iou": 0.51822387496614,
196
+ "eval_loss": 0.7638939619064331,
197
+ "eval_runtime": 4.4461,
198
+ "eval_samples_per_second": 22.491,
199
+ "eval_steps_per_second": 2.924,
200
+ "step": 500
201
+ }
202
+ ],
203
+ "logging_steps": 25,
204
+ "max_steps": 1000,
205
+ "num_input_tokens_seen": 0,
206
+ "num_train_epochs": 10,
207
+ "save_steps": 500,
208
+ "stateful_callbacks": {
209
+ "TrainerControl": {
210
+ "args": {
211
+ "should_epoch_stop": false,
212
+ "should_evaluate": false,
213
+ "should_log": false,
214
+ "should_save": true,
215
+ "should_training_stop": false
216
+ },
217
+ "attributes": {}
218
+ }
219
+ },
220
+ "total_flos": 0.0,
221
+ "train_batch_size": 8,
222
+ "trial_name": null,
223
+ "trial_params": null
224
+ }
checkpoint-500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04792f1700f54edbe0f85471cd1c5824c640ef6ced03e54f94ffcf8b7d12faea
3
+ size 5201
checkpoint-600/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:169a17d2842d3d3c0f802d1829bdc26d599f19ce1692e6bf9813ae3b82137d5e
3
+ size 124209532
checkpoint-600/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9d89ac787895a35f31e00f49fbf66b6061fa48dcd8bc1dd0250d5fb9db001f6
3
+ size 248356555
checkpoint-600/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1255b82ec9f462fef6f8c2f4a8ee876c4b6d998a53c99546a4eddcc1aabaefc7
3
+ size 14645
checkpoint-600/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ccf18be5c41ff4265589382c300ac27c252a672e83a95cf1c2983a2a0b62049
3
+ size 1465
checkpoint-600/trainer_state.json ADDED
@@ -0,0 +1,262 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 600,
3
+ "best_metric": 0.7196315708151912,
4
+ "best_model_checkpoint": "./kvasir-unet/checkpoint-600",
5
+ "epoch": 6.0,
6
+ "eval_steps": 500,
7
+ "global_step": 600,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.25,
14
+ "grad_norm": 1.4807058572769165,
15
+ "learning_rate": 9.76e-05,
16
+ "loss": 1.2226847076416016,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.5,
21
+ "grad_norm": 1.6061848402023315,
22
+ "learning_rate": 9.51e-05,
23
+ "loss": 1.1226789093017577,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.75,
28
+ "grad_norm": 1.5647845268249512,
29
+ "learning_rate": 9.260000000000001e-05,
30
+ "loss": 1.0822339630126954,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 1.0,
35
+ "grad_norm": 4.204342365264893,
36
+ "learning_rate": 9.010000000000001e-05,
37
+ "loss": 1.00395751953125,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 1.0,
42
+ "eval_dice": 0.4649257596145264,
43
+ "eval_iou": 0.3028688409242743,
44
+ "eval_loss": 1.0397001504898071,
45
+ "eval_runtime": 4.527,
46
+ "eval_samples_per_second": 22.09,
47
+ "eval_steps_per_second": 2.872,
48
+ "step": 100
49
+ },
50
+ {
51
+ "epoch": 1.25,
52
+ "grad_norm": 2.125596284866333,
53
+ "learning_rate": 8.76e-05,
54
+ "loss": 0.9969179534912109,
55
+ "step": 125
56
+ },
57
+ {
58
+ "epoch": 1.5,
59
+ "grad_norm": 2.8102705478668213,
60
+ "learning_rate": 8.510000000000001e-05,
61
+ "loss": 0.9738675689697266,
62
+ "step": 150
63
+ },
64
+ {
65
+ "epoch": 1.75,
66
+ "grad_norm": 2.3319432735443115,
67
+ "learning_rate": 8.26e-05,
68
+ "loss": 0.9797834777832031,
69
+ "step": 175
70
+ },
71
+ {
72
+ "epoch": 2.0,
73
+ "grad_norm": 3.546826124191284,
74
+ "learning_rate": 8.010000000000001e-05,
75
+ "loss": 0.936845932006836,
76
+ "step": 200
77
+ },
78
+ {
79
+ "epoch": 2.0,
80
+ "eval_dice": 0.5913469032255761,
81
+ "eval_iou": 0.41979614051111847,
82
+ "eval_loss": 0.9874243140220642,
83
+ "eval_runtime": 4.4621,
84
+ "eval_samples_per_second": 22.411,
85
+ "eval_steps_per_second": 2.913,
86
+ "step": 200
87
+ },
88
+ {
89
+ "epoch": 2.25,
90
+ "grad_norm": 4.708238124847412,
91
+ "learning_rate": 7.76e-05,
92
+ "loss": 0.9107416534423828,
93
+ "step": 225
94
+ },
95
+ {
96
+ "epoch": 2.5,
97
+ "grad_norm": 1.9260565042495728,
98
+ "learning_rate": 7.510000000000001e-05,
99
+ "loss": 0.9003370666503906,
100
+ "step": 250
101
+ },
102
+ {
103
+ "epoch": 2.75,
104
+ "grad_norm": 2.7347285747528076,
105
+ "learning_rate": 7.26e-05,
106
+ "loss": 0.8682646942138672,
107
+ "step": 275
108
+ },
109
+ {
110
+ "epoch": 3.0,
111
+ "grad_norm": 1.9796888828277588,
112
+ "learning_rate": 7.01e-05,
113
+ "loss": 0.8491334533691406,
114
+ "step": 300
115
+ },
116
+ {
117
+ "epoch": 3.0,
118
+ "eval_dice": 0.6287870060272235,
119
+ "eval_iou": 0.45856278594402466,
120
+ "eval_loss": 0.880113422870636,
121
+ "eval_runtime": 4.6294,
122
+ "eval_samples_per_second": 21.601,
123
+ "eval_steps_per_second": 2.808,
124
+ "step": 300
125
+ },
126
+ {
127
+ "epoch": 3.25,
128
+ "grad_norm": 4.970420837402344,
129
+ "learning_rate": 6.76e-05,
130
+ "loss": 0.8522591400146484,
131
+ "step": 325
132
+ },
133
+ {
134
+ "epoch": 3.5,
135
+ "grad_norm": 2.374986171722412,
136
+ "learning_rate": 6.510000000000001e-05,
137
+ "loss": 0.7948677825927735,
138
+ "step": 350
139
+ },
140
+ {
141
+ "epoch": 3.75,
142
+ "grad_norm": 5.10560417175293,
143
+ "learning_rate": 6.26e-05,
144
+ "loss": 0.8352813720703125,
145
+ "step": 375
146
+ },
147
+ {
148
+ "epoch": 4.0,
149
+ "grad_norm": 4.5725274085998535,
150
+ "learning_rate": 6.0100000000000004e-05,
151
+ "loss": 0.7697490692138672,
152
+ "step": 400
153
+ },
154
+ {
155
+ "epoch": 4.0,
156
+ "eval_dice": 0.6448364361669014,
157
+ "eval_iou": 0.4758367704940324,
158
+ "eval_loss": 0.8459730744361877,
159
+ "eval_runtime": 4.4418,
160
+ "eval_samples_per_second": 22.514,
161
+ "eval_steps_per_second": 2.927,
162
+ "step": 400
163
+ },
164
+ {
165
+ "epoch": 4.25,
166
+ "grad_norm": 7.802286148071289,
167
+ "learning_rate": 5.76e-05,
168
+ "loss": 0.7570986175537109,
169
+ "step": 425
170
+ },
171
+ {
172
+ "epoch": 4.5,
173
+ "grad_norm": 7.580319881439209,
174
+ "learning_rate": 5.5100000000000004e-05,
175
+ "loss": 0.763985366821289,
176
+ "step": 450
177
+ },
178
+ {
179
+ "epoch": 4.75,
180
+ "grad_norm": 3.051490068435669,
181
+ "learning_rate": 5.2600000000000005e-05,
182
+ "loss": 0.7334027862548829,
183
+ "step": 475
184
+ },
185
+ {
186
+ "epoch": 5.0,
187
+ "grad_norm": 3.4101736545562744,
188
+ "learning_rate": 5.0100000000000005e-05,
189
+ "loss": 0.7454586791992187,
190
+ "step": 500
191
+ },
192
+ {
193
+ "epoch": 5.0,
194
+ "eval_dice": 0.6826710522273053,
195
+ "eval_iou": 0.51822387496614,
196
+ "eval_loss": 0.7638939619064331,
197
+ "eval_runtime": 4.4461,
198
+ "eval_samples_per_second": 22.491,
199
+ "eval_steps_per_second": 2.924,
200
+ "step": 500
201
+ },
202
+ {
203
+ "epoch": 5.25,
204
+ "grad_norm": 9.261576652526855,
205
+ "learning_rate": 4.76e-05,
206
+ "loss": 0.7240135955810547,
207
+ "step": 525
208
+ },
209
+ {
210
+ "epoch": 5.5,
211
+ "grad_norm": 3.081444501876831,
212
+ "learning_rate": 4.5100000000000005e-05,
213
+ "loss": 0.7151520538330078,
214
+ "step": 550
215
+ },
216
+ {
217
+ "epoch": 5.75,
218
+ "grad_norm": 4.129289150238037,
219
+ "learning_rate": 4.26e-05,
220
+ "loss": 0.6827881622314453,
221
+ "step": 575
222
+ },
223
+ {
224
+ "epoch": 6.0,
225
+ "grad_norm": 3.271570920944214,
226
+ "learning_rate": 4.0100000000000006e-05,
227
+ "loss": 0.6640605926513672,
228
+ "step": 600
229
+ },
230
+ {
231
+ "epoch": 6.0,
232
+ "eval_dice": 0.7196315708151912,
233
+ "eval_iou": 0.562050553478283,
234
+ "eval_loss": 0.7080976366996765,
235
+ "eval_runtime": 4.5035,
236
+ "eval_samples_per_second": 22.205,
237
+ "eval_steps_per_second": 2.887,
238
+ "step": 600
239
+ }
240
+ ],
241
+ "logging_steps": 25,
242
+ "max_steps": 1000,
243
+ "num_input_tokens_seen": 0,
244
+ "num_train_epochs": 10,
245
+ "save_steps": 500,
246
+ "stateful_callbacks": {
247
+ "TrainerControl": {
248
+ "args": {
249
+ "should_epoch_stop": false,
250
+ "should_evaluate": false,
251
+ "should_log": false,
252
+ "should_save": true,
253
+ "should_training_stop": false
254
+ },
255
+ "attributes": {}
256
+ }
257
+ },
258
+ "total_flos": 0.0,
259
+ "train_batch_size": 8,
260
+ "trial_name": null,
261
+ "trial_params": null
262
+ }
checkpoint-600/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04792f1700f54edbe0f85471cd1c5824c640ef6ced03e54f94ffcf8b7d12faea
3
+ size 5201
checkpoint-700/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e537ad3d7de71931337d69a359aa74f1fac8b7bf5452ad016686db0f6e29f5a
3
+ size 124209532
checkpoint-700/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ece3f70695e6ab6cc0637bd72fefc15e82a1b5e0b78917cf69b6b5efa4b00c39
3
+ size 248356555
checkpoint-700/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20ea3a198ff666cb4ace1c684b598fe43fc7c3c276b83efc553a1b787e12a304
3
+ size 14645
checkpoint-700/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b43ff283015c76f51472644d93941dba2258d316a76706a538d778d318f06460
3
+ size 1465
checkpoint-700/trainer_state.json ADDED
@@ -0,0 +1,300 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 700,
3
+ "best_metric": 0.7246417440175368,
4
+ "best_model_checkpoint": "./kvasir-unet/checkpoint-700",
5
+ "epoch": 7.0,
6
+ "eval_steps": 500,
7
+ "global_step": 700,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.25,
14
+ "grad_norm": 1.4807058572769165,
15
+ "learning_rate": 9.76e-05,
16
+ "loss": 1.2226847076416016,
17
+ "step": 25
18
+ },
19
+ {
20
+ "epoch": 0.5,
21
+ "grad_norm": 1.6061848402023315,
22
+ "learning_rate": 9.51e-05,
23
+ "loss": 1.1226789093017577,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.75,
28
+ "grad_norm": 1.5647845268249512,
29
+ "learning_rate": 9.260000000000001e-05,
30
+ "loss": 1.0822339630126954,
31
+ "step": 75
32
+ },
33
+ {
34
+ "epoch": 1.0,
35
+ "grad_norm": 4.204342365264893,
36
+ "learning_rate": 9.010000000000001e-05,
37
+ "loss": 1.00395751953125,
38
+ "step": 100
39
+ },
40
+ {
41
+ "epoch": 1.0,
42
+ "eval_dice": 0.4649257596145264,
43
+ "eval_iou": 0.3028688409242743,
44
+ "eval_loss": 1.0397001504898071,
45
+ "eval_runtime": 4.527,
46
+ "eval_samples_per_second": 22.09,
47
+ "eval_steps_per_second": 2.872,
48
+ "step": 100
49
+ },
50
+ {
51
+ "epoch": 1.25,
52
+ "grad_norm": 2.125596284866333,
53
+ "learning_rate": 8.76e-05,
54
+ "loss": 0.9969179534912109,
55
+ "step": 125
56
+ },
57
+ {
58
+ "epoch": 1.5,
59
+ "grad_norm": 2.8102705478668213,
60
+ "learning_rate": 8.510000000000001e-05,
61
+ "loss": 0.9738675689697266,
62
+ "step": 150
63
+ },
64
+ {
65
+ "epoch": 1.75,
66
+ "grad_norm": 2.3319432735443115,
67
+ "learning_rate": 8.26e-05,
68
+ "loss": 0.9797834777832031,
69
+ "step": 175
70
+ },
71
+ {
72
+ "epoch": 2.0,
73
+ "grad_norm": 3.546826124191284,
74
+ "learning_rate": 8.010000000000001e-05,
75
+ "loss": 0.936845932006836,
76
+ "step": 200
77
+ },
78
+ {
79
+ "epoch": 2.0,
80
+ "eval_dice": 0.5913469032255761,
81
+ "eval_iou": 0.41979614051111847,
82
+ "eval_loss": 0.9874243140220642,
83
+ "eval_runtime": 4.4621,
84
+ "eval_samples_per_second": 22.411,
85
+ "eval_steps_per_second": 2.913,
86
+ "step": 200
87
+ },
88
+ {
89
+ "epoch": 2.25,
90
+ "grad_norm": 4.708238124847412,
91
+ "learning_rate": 7.76e-05,
92
+ "loss": 0.9107416534423828,
93
+ "step": 225
94
+ },
95
+ {
96
+ "epoch": 2.5,
97
+ "grad_norm": 1.9260565042495728,
98
+ "learning_rate": 7.510000000000001e-05,
99
+ "loss": 0.9003370666503906,
100
+ "step": 250
101
+ },
102
+ {
103
+ "epoch": 2.75,
104
+ "grad_norm": 2.7347285747528076,
105
+ "learning_rate": 7.26e-05,
106
+ "loss": 0.8682646942138672,
107
+ "step": 275
108
+ },
109
+ {
110
+ "epoch": 3.0,
111
+ "grad_norm": 1.9796888828277588,
112
+ "learning_rate": 7.01e-05,
113
+ "loss": 0.8491334533691406,
114
+ "step": 300
115
+ },
116
+ {
117
+ "epoch": 3.0,
118
+ "eval_dice": 0.6287870060272235,
119
+ "eval_iou": 0.45856278594402466,
120
+ "eval_loss": 0.880113422870636,
121
+ "eval_runtime": 4.6294,
122
+ "eval_samples_per_second": 21.601,
123
+ "eval_steps_per_second": 2.808,
124
+ "step": 300
125
+ },
126
+ {
127
+ "epoch": 3.25,
128
+ "grad_norm": 4.970420837402344,
129
+ "learning_rate": 6.76e-05,
130
+ "loss": 0.8522591400146484,
131
+ "step": 325
132
+ },
133
+ {
134
+ "epoch": 3.5,
135
+ "grad_norm": 2.374986171722412,
136
+ "learning_rate": 6.510000000000001e-05,
137
+ "loss": 0.7948677825927735,
138
+ "step": 350
139
+ },
140
+ {
141
+ "epoch": 3.75,
142
+ "grad_norm": 5.10560417175293,
143
+ "learning_rate": 6.26e-05,
144
+ "loss": 0.8352813720703125,
145
+ "step": 375
146
+ },
147
+ {
148
+ "epoch": 4.0,
149
+ "grad_norm": 4.5725274085998535,
150
+ "learning_rate": 6.0100000000000004e-05,
151
+ "loss": 0.7697490692138672,
152
+ "step": 400
153
+ },
154
+ {
155
+ "epoch": 4.0,
156
+ "eval_dice": 0.6448364361669014,
157
+ "eval_iou": 0.4758367704940324,
158
+ "eval_loss": 0.8459730744361877,
159
+ "eval_runtime": 4.4418,
160
+ "eval_samples_per_second": 22.514,
161
+ "eval_steps_per_second": 2.927,
162
+ "step": 400
163
+ },
164
+ {
165
+ "epoch": 4.25,
166
+ "grad_norm": 7.802286148071289,
167
+ "learning_rate": 5.76e-05,
168
+ "loss": 0.7570986175537109,
169
+ "step": 425
170
+ },
171
+ {
172
+ "epoch": 4.5,
173
+ "grad_norm": 7.580319881439209,
174
+ "learning_rate": 5.5100000000000004e-05,
175
+ "loss": 0.763985366821289,
176
+ "step": 450
177
+ },
178
+ {
179
+ "epoch": 4.75,
180
+ "grad_norm": 3.051490068435669,
181
+ "learning_rate": 5.2600000000000005e-05,
182
+ "loss": 0.7334027862548829,
183
+ "step": 475
184
+ },
185
+ {
186
+ "epoch": 5.0,
187
+ "grad_norm": 3.4101736545562744,
188
+ "learning_rate": 5.0100000000000005e-05,
189
+ "loss": 0.7454586791992187,
190
+ "step": 500
191
+ },
192
+ {
193
+ "epoch": 5.0,
194
+ "eval_dice": 0.6826710522273053,
195
+ "eval_iou": 0.51822387496614,
196
+ "eval_loss": 0.7638939619064331,
197
+ "eval_runtime": 4.4461,
198
+ "eval_samples_per_second": 22.491,
199
+ "eval_steps_per_second": 2.924,
200
+ "step": 500
201
+ },
202
+ {
203
+ "epoch": 5.25,
204
+ "grad_norm": 9.261576652526855,
205
+ "learning_rate": 4.76e-05,
206
+ "loss": 0.7240135955810547,
207
+ "step": 525
208
+ },
209
+ {
210
+ "epoch": 5.5,
211
+ "grad_norm": 3.081444501876831,
212
+ "learning_rate": 4.5100000000000005e-05,
213
+ "loss": 0.7151520538330078,
214
+ "step": 550
215
+ },
216
+ {
217
+ "epoch": 5.75,
218
+ "grad_norm": 4.129289150238037,
219
+ "learning_rate": 4.26e-05,
220
+ "loss": 0.6827881622314453,
221
+ "step": 575
222
+ },
223
+ {
224
+ "epoch": 6.0,
225
+ "grad_norm": 3.271570920944214,
226
+ "learning_rate": 4.0100000000000006e-05,
227
+ "loss": 0.6640605926513672,
228
+ "step": 600
229
+ },
230
+ {
231
+ "epoch": 6.0,
232
+ "eval_dice": 0.7196315708151912,
233
+ "eval_iou": 0.562050553478283,
234
+ "eval_loss": 0.7080976366996765,
235
+ "eval_runtime": 4.5035,
236
+ "eval_samples_per_second": 22.205,
237
+ "eval_steps_per_second": 2.887,
238
+ "step": 600
239
+ },
240
+ {
241
+ "epoch": 6.25,
242
+ "grad_norm": 3.007444143295288,
243
+ "learning_rate": 3.76e-05,
244
+ "loss": 0.6774704742431641,
245
+ "step": 625
246
+ },
247
+ {
248
+ "epoch": 6.5,
249
+ "grad_norm": 6.9112420082092285,
250
+ "learning_rate": 3.51e-05,
251
+ "loss": 0.6406173706054688,
252
+ "step": 650
253
+ },
254
+ {
255
+ "epoch": 6.75,
256
+ "grad_norm": 3.0203912258148193,
257
+ "learning_rate": 3.26e-05,
258
+ "loss": 0.6287059783935547,
259
+ "step": 675
260
+ },
261
+ {
262
+ "epoch": 7.0,
263
+ "grad_norm": 3.216691732406616,
264
+ "learning_rate": 3.01e-05,
265
+ "loss": 0.647440185546875,
266
+ "step": 700
267
+ },
268
+ {
269
+ "epoch": 7.0,
270
+ "eval_dice": 0.7246417440175368,
271
+ "eval_iou": 0.5681869886775344,
272
+ "eval_loss": 0.691881000995636,
273
+ "eval_runtime": 4.5083,
274
+ "eval_samples_per_second": 22.181,
275
+ "eval_steps_per_second": 2.884,
276
+ "step": 700
277
+ }
278
+ ],
279
+ "logging_steps": 25,
280
+ "max_steps": 1000,
281
+ "num_input_tokens_seen": 0,
282
+ "num_train_epochs": 10,
283
+ "save_steps": 500,
284
+ "stateful_callbacks": {
285
+ "TrainerControl": {
286
+ "args": {
287
+ "should_epoch_stop": false,
288
+ "should_evaluate": false,
289
+ "should_log": false,
290
+ "should_save": true,
291
+ "should_training_stop": false
292
+ },
293
+ "attributes": {}
294
+ }
295
+ },
296
+ "total_flos": 0.0,
297
+ "train_batch_size": 8,
298
+ "trial_name": null,
299
+ "trial_params": null
300
+ }
checkpoint-700/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04792f1700f54edbe0f85471cd1c5824c640ef6ced03e54f94ffcf8b7d12faea
3
+ size 5201
checkpoint-800/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6fcf73cb27f18b31a9ba7d20e4092a1b90723a5fa526c927770ad1f2ab5d63d
3
+ size 124209532