Hugging Face's logo

BTX24
/
levit_128.fb_dist_in1k-finetuned-stroke-binary

Image Classification
Transformers
TensorBoard
Safetensors
timm_wrapper
Generated from Trainer
Model card Files Metrics Community
BTX24 commited on
Commit
a7c0627
·
verified ·
1 Parent(s): fa37af7

Training in progress, step 100

Browse files
Files changed (6) hide show
  1. all_results.json +8 -0
  2. model.safetensors +1 -1
  3. runs/Mar17_13-43-41_d7c13a14123e/events.out.tfevents.1742219035.d7c13a14123e.3775.9 +3 -0
  4. train_results.json +8 -0
  5. trainer_state.json +276 -0
  6. training_args.bin +1 -1
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.558139534883721,
3
+ "total_flos": 2.1868569076064256e+17,
4
+ "train_loss": 0.6519456799825033,
5
+ "train_runtime": 297.561,
6
+ "train_samples_per_second": 624.033,
7
+ "train_steps_per_second": 19.478
8
+ }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4c9b32d51901e0d7cf59ea04fae4aa8ae565da4abe2be03541fcb763fa36a952
3
  size 34019112
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26726aa05dfda67f8375f3f58a12bbad60f9c19af6ca49f8d4d570d897427a24
3
  size 34019112
runs/Mar17_13-43-41_d7c13a14123e/events.out.tfevents.1742219035.d7c13a14123e.3775.9 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37fd49c1231165cb796ce63be3c97c33766be5541efd7ee00c0ff39e5320f92c
3
+ size 6194
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.558139534883721,
3
+ "total_flos": 2.1868569076064256e+17,
4
+ "train_loss": 0.6519456799825033,
5
+ "train_runtime": 297.561,
6
+ "train_samples_per_second": 624.033,
7
+ "train_steps_per_second": 19.478
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,276 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7092775390280552,
3
+ "best_model_checkpoint": "levit_128.fb_dist_in1k-finetuned-stroke-binary/checkpoint-600",
4
+ "epoch": 5.558139534883721,
5
+ "eval_steps": 100,
6
+ "global_step": 900,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.31007751937984496,
13
+ "grad_norm": 175.24969482421875,
14
+ "learning_rate": 1.413793103448276e-06,
15
+ "loss": 0.6951,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.6201550387596899,
20
+ "grad_norm": 146.73788452148438,
21
+ "learning_rate": 3.137931034482759e-06,
22
+ "loss": 0.695,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.6201550387596899,
27
+ "eval_accuracy": 0.5364088647670737,
28
+ "eval_f1": 0.5420516079700721,
29
+ "eval_loss": NaN,
30
+ "eval_precision": 0.5540979766391916,
31
+ "eval_recall": 0.5364088647670737,
32
+ "eval_runtime": 6.9382,
33
+ "eval_samples_per_second": 318.669,
34
+ "eval_steps_per_second": 39.924,
35
+ "step": 100
36
+ },
37
+ {
38
+ "epoch": 0.9302325581395349,
39
+ "grad_norm": 127.28129577636719,
40
+ "learning_rate": 4.793103448275862e-06,
41
+ "loss": 0.6942,
42
+ "step": 150
43
+ },
44
+ {
45
+ "epoch": 1.235658914728682,
46
+ "grad_norm": 751.1776123046875,
47
+ "learning_rate": 6.482758620689655e-06,
48
+ "loss": 0.6804,
49
+ "step": 200
50
+ },
51
+ {
52
+ "epoch": 1.235658914728682,
53
+ "eval_accuracy": 0.5798281320669381,
54
+ "eval_f1": 0.5832968695954936,
55
+ "eval_loss": NaN,
56
+ "eval_precision": 0.6277845389250567,
57
+ "eval_recall": 0.5798281320669381,
58
+ "eval_runtime": 6.9012,
59
+ "eval_samples_per_second": 320.38,
60
+ "eval_steps_per_second": 40.138,
61
+ "step": 200
62
+ },
63
+ {
64
+ "epoch": 1.5457364341085271,
65
+ "grad_norm": 141.23553466796875,
66
+ "learning_rate": 8.206896551724138e-06,
67
+ "loss": 0.6896,
68
+ "step": 250
69
+ },
70
+ {
71
+ "epoch": 1.8558139534883722,
72
+ "grad_norm": 39.63947296142578,
73
+ "learning_rate": 9.931034482758622e-06,
74
+ "loss": 0.6821,
75
+ "step": 300
76
+ },
77
+ {
78
+ "epoch": 1.8558139534883722,
79
+ "eval_accuracy": 0.6232473993668024,
80
+ "eval_f1": 0.6279682812403823,
81
+ "eval_loss": NaN,
82
+ "eval_precision": 0.6412845278993262,
83
+ "eval_recall": 0.6232473993668024,
84
+ "eval_runtime": 6.5774,
85
+ "eval_samples_per_second": 336.153,
86
+ "eval_steps_per_second": 42.114,
87
+ "step": 300
88
+ },
89
+ {
90
+ "epoch": 2.1612403100775195,
91
+ "grad_norm": 176.00396728515625,
92
+ "learning_rate": 1.1655172413793106e-05,
93
+ "loss": 0.6687,
94
+ "step": 350
95
+ },
96
+ {
97
+ "epoch": 2.471317829457364,
98
+ "grad_norm": 245.2242889404297,
99
+ "learning_rate": 1.3379310344827587e-05,
100
+ "loss": 0.6726,
101
+ "step": 400
102
+ },
103
+ {
104
+ "epoch": 2.471317829457364,
105
+ "eval_accuracy": 0.6671189507010402,
106
+ "eval_f1": 0.6711377069586024,
107
+ "eval_loss": NaN,
108
+ "eval_precision": 0.6829477528090242,
109
+ "eval_recall": 0.6671189507010402,
110
+ "eval_runtime": 6.8482,
111
+ "eval_samples_per_second": 322.86,
112
+ "eval_steps_per_second": 40.449,
113
+ "step": 400
114
+ },
115
+ {
116
+ "epoch": 2.781395348837209,
117
+ "grad_norm": 17503.568359375,
118
+ "learning_rate": 1.5068965517241381e-05,
119
+ "loss": 0.6682,
120
+ "step": 450
121
+ },
122
+ {
123
+ "epoch": 3.0868217054263565,
124
+ "grad_norm": 69.87994384765625,
125
+ "learning_rate": 1.6793103448275862e-05,
126
+ "loss": 0.6546,
127
+ "step": 500
128
+ },
129
+ {
130
+ "epoch": 3.0868217054263565,
131
+ "eval_accuracy": 0.70239710538218,
132
+ "eval_f1": 0.7020627244507841,
133
+ "eval_loss": NaN,
134
+ "eval_precision": 0.7017573042623244,
135
+ "eval_recall": 0.70239710538218,
136
+ "eval_runtime": 6.8785,
137
+ "eval_samples_per_second": 321.438,
138
+ "eval_steps_per_second": 40.271,
139
+ "step": 500
140
+ },
141
+ {
142
+ "epoch": 3.3968992248062015,
143
+ "grad_norm": 101.73799133300781,
144
+ "learning_rate": 1.8517241379310348e-05,
145
+ "loss": 0.6508,
146
+ "step": 550
147
+ },
148
+ {
149
+ "epoch": 3.7069767441860466,
150
+ "grad_norm": 103.15814971923828,
151
+ "learning_rate": 1.9999934702491228e-05,
152
+ "loss": 0.647,
153
+ "step": 600
154
+ },
155
+ {
156
+ "epoch": 3.7069767441860466,
157
+ "eval_accuracy": 0.7064676616915423,
158
+ "eval_f1": 0.7092775390280552,
159
+ "eval_loss": NaN,
160
+ "eval_precision": 0.7158812934752514,
161
+ "eval_recall": 0.7064676616915423,
162
+ "eval_runtime": 6.9098,
163
+ "eval_samples_per_second": 319.98,
164
+ "eval_steps_per_second": 40.088,
165
+ "step": 600
166
+ },
167
+ {
168
+ "epoch": 4.0124031007751935,
169
+ "grad_norm": 169.00445556640625,
170
+ "learning_rate": 1.9994312394497127e-05,
171
+ "loss": 0.6255,
172
+ "step": 650
173
+ },
174
+ {
175
+ "epoch": 4.322480620155039,
176
+ "grad_norm": 8452.361328125,
177
+ "learning_rate": 1.997962682133918e-05,
178
+ "loss": 0.6263,
179
+ "step": 700
180
+ },
181
+ {
182
+ "epoch": 4.322480620155039,
183
+ "eval_accuracy": 0.6956128448665763,
184
+ "eval_f1": 0.6991433808448707,
185
+ "eval_loss": NaN,
186
+ "eval_precision": 0.709618757975013,
187
+ "eval_recall": 0.6956128448665763,
188
+ "eval_runtime": 6.9194,
189
+ "eval_samples_per_second": 319.537,
190
+ "eval_steps_per_second": 40.032,
191
+ "step": 700
192
+ },
193
+ {
194
+ "epoch": 4.632558139534884,
195
+ "grad_norm": 53.048866271972656,
196
+ "learning_rate": 1.9955891300516223e-05,
197
+ "loss": 0.6144,
198
+ "step": 750
199
+ },
200
+ {
201
+ "epoch": 4.942635658914728,
202
+ "grad_norm": 66.52362823486328,
203
+ "learning_rate": 1.9923127356402373e-05,
204
+ "loss": 0.6112,
205
+ "step": 800
206
+ },
207
+ {
208
+ "epoch": 4.942635658914728,
209
+ "eval_accuracy": 0.6766169154228856,
210
+ "eval_f1": 0.6806907989098179,
211
+ "eval_loss": NaN,
212
+ "eval_precision": 0.7122760837780127,
213
+ "eval_recall": 0.6766169154228856,
214
+ "eval_runtime": 6.6311,
215
+ "eval_samples_per_second": 333.428,
216
+ "eval_steps_per_second": 41.773,
217
+ "step": 800
218
+ },
219
+ {
220
+ "epoch": 5.248062015503876,
221
+ "grad_norm": 39.8150634765625,
222
+ "learning_rate": 1.9881364700727827e-05,
223
+ "loss": 0.5889,
224
+ "step": 850
225
+ },
226
+ {
227
+ "epoch": 5.558139534883721,
228
+ "grad_norm": 156.32704162597656,
229
+ "learning_rate": 1.9830641205634987e-05,
230
+ "loss": 0.5704,
231
+ "step": 900
232
+ },
233
+ {
234
+ "epoch": 5.558139534883721,
235
+ "eval_accuracy": 0.6974219810040706,
236
+ "eval_f1": 0.7013769772020675,
237
+ "eval_loss": NaN,
238
+ "eval_precision": 0.7297910740241028,
239
+ "eval_recall": 0.6974219810040706,
240
+ "eval_runtime": 6.7688,
241
+ "eval_samples_per_second": 326.644,
242
+ "eval_steps_per_second": 40.923,
243
+ "step": 900
244
+ },
245
+ {
246
+ "epoch": 5.558139534883721,
247
+ "step": 900,
248
+ "total_flos": 2.1868569076064256e+17,
249
+ "train_loss": 0.6519456799825033,
250
+ "train_runtime": 297.561,
251
+ "train_samples_per_second": 624.033,
252
+ "train_steps_per_second": 19.478
253
+ }
254
+ ],
255
+ "logging_steps": 50,
256
+ "max_steps": 5796,
257
+ "num_input_tokens_seen": 0,
258
+ "num_train_epochs": 36,
259
+ "save_steps": 100,
260
+ "stateful_callbacks": {
261
+ "TrainerControl": {
262
+ "args": {
263
+ "should_epoch_stop": false,
264
+ "should_evaluate": false,
265
+ "should_log": false,
266
+ "should_save": true,
267
+ "should_training_stop": true
268
+ },
269
+ "attributes": {}
270
+ }
271
+ },
272
+ "total_flos": 2.1868569076064256e+17,
273
+ "train_batch_size": 8,
274
+ "trial_name": null,
275
+ "trial_params": null
276
+ }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1ab6b7af41ae1b2821d9aba1144186612cf22cc84d83b5ee612bc5ceab0e174f
3
  size 5368
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a468310e8d3f5b7bb70bf943e579a95d88353826db4e1833b1651cb332b4393c
3
  size 5368