jatinkanduri commited on
Commit
42861d2
·
verified ·
1 Parent(s): 6d91e8a

Upload folder using huggingface_hub

Browse files
checkpoint-198/config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/swin-base-patch4-window7-224",
3
+ "architectures": [
4
+ "SwinForImageClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.0,
7
+ "depths": [
8
+ 2,
9
+ 2,
10
+ 18,
11
+ 2
12
+ ],
13
+ "drop_path_rate": 0.1,
14
+ "embed_dim": 128,
15
+ "encoder_stride": 32,
16
+ "hidden_act": "gelu",
17
+ "hidden_dropout_prob": 0.0,
18
+ "hidden_size": 1024,
19
+ "image_size": 224,
20
+ "initializer_range": 0.02,
21
+ "id2label": {
22
+ "0": "defective",
23
+ "1": "good"
24
+ },
25
+ "label2id": {
26
+ "defective": 0,
27
+ "good": 1
28
+ },
29
+ "layer_norm_eps": 1e-05,
30
+ "mlp_ratio": 4.0,
31
+ "model_type": "swin",
32
+ "num_channels": 3,
33
+ "num_heads": [
34
+ 4,
35
+ 8,
36
+ 16,
37
+ 32
38
+ ],
39
+ "num_layers": 4,
40
+ "out_features": [
41
+ "stage4"
42
+ ],
43
+ "out_indices": [
44
+ 4
45
+ ],
46
+ "patch_size": 4,
47
+ "path_norm": true,
48
+ "problem_type": "single_label_classification",
49
+ "qkv_bias": true,
50
+ "stage_names": [
51
+ "stem",
52
+ "stage1",
53
+ "stage2",
54
+ "stage3",
55
+ "stage4"
56
+ ],
57
+ "torch_dtype": "float32",
58
+ "transformers_version": "4.47.1",
59
+ "use_absolute_embeddings": false,
60
+ "window_size": 7
61
+ }
checkpoint-198/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99606d1b17c3bb84a963911cf5e0a1b59d20aca5bbb7cd24d88d1310ef2a9e93
3
+ size 347498816
checkpoint-198/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a085917f789eb14431140c8009f8085613d72869a0ad736284980020f76b0d37
3
+ size 694317645
checkpoint-198/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8db2377cde69008c7654229ef6a78312b6f9d819ee699ba1a7e6832fa98f3fb0
3
+ size 14244
checkpoint-198/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7e049f0ed12ef74793e7faff81beebf1a693b76504356603bc2c594206b1560
3
+ size 1064
checkpoint-198/trainer_state.json ADDED
@@ -0,0 +1,283 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8301314459049545,
3
+ "best_model_checkpoint": "./swin_final_model/checkpoint-88",
4
+ "epoch": 9.0,
5
+ "eval_steps": 500,
6
+ "global_step": 198,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.45454545454545453,
13
+ "grad_norm": 5.4057230949401855,
14
+ "learning_rate": 4.993630286525634e-05,
15
+ "loss": 0.1972,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.9090909090909091,
20
+ "grad_norm": 3.5544466972351074,
21
+ "learning_rate": 4.9745536047023324e-05,
22
+ "loss": 0.1455,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 1.0,
27
+ "eval_accuracy": 0.6785714285714286,
28
+ "eval_f1": 0.40425531914893614,
29
+ "eval_loss": 0.1654156595468521,
30
+ "eval_precision": 0.3392857142857143,
31
+ "eval_recall": 0.5,
32
+ "eval_runtime": 2.0657,
33
+ "eval_samples_per_second": 27.11,
34
+ "eval_steps_per_second": 1.936,
35
+ "step": 22
36
+ },
37
+ {
38
+ "epoch": 1.3636363636363638,
39
+ "grad_norm": 34.79863357543945,
40
+ "learning_rate": 4.942867164927899e-05,
41
+ "loss": 0.1528,
42
+ "step": 30
43
+ },
44
+ {
45
+ "epoch": 1.8181818181818183,
46
+ "grad_norm": 3.0228710174560547,
47
+ "learning_rate": 4.898732434036244e-05,
48
+ "loss": 0.1415,
49
+ "step": 40
50
+ },
51
+ {
52
+ "epoch": 2.0,
53
+ "eval_accuracy": 0.6785714285714286,
54
+ "eval_f1": 0.6199095022624435,
55
+ "eval_loss": 0.15097904205322266,
56
+ "eval_precision": 0.625,
57
+ "eval_recall": 0.6169590643274854,
58
+ "eval_runtime": 2.0117,
59
+ "eval_samples_per_second": 27.838,
60
+ "eval_steps_per_second": 1.988,
61
+ "step": 44
62
+ },
63
+ {
64
+ "epoch": 2.2727272727272725,
65
+ "grad_norm": 4.495301246643066,
66
+ "learning_rate": 4.842374312499405e-05,
67
+ "loss": 0.1288,
68
+ "step": 50
69
+ },
70
+ {
71
+ "epoch": 2.7272727272727275,
72
+ "grad_norm": 2.2047574520111084,
73
+ "learning_rate": 4.774079988386296e-05,
74
+ "loss": 0.1213,
75
+ "step": 60
76
+ },
77
+ {
78
+ "epoch": 3.0,
79
+ "eval_accuracy": 0.7142857142857143,
80
+ "eval_f1": 0.7005347593582887,
81
+ "eval_loss": 0.1337863653898239,
82
+ "eval_precision": 0.7025641025641025,
83
+ "eval_recall": 0.7309941520467836,
84
+ "eval_runtime": 2.1054,
85
+ "eval_samples_per_second": 26.599,
86
+ "eval_steps_per_second": 1.9,
87
+ "step": 66
88
+ },
89
+ {
90
+ "epoch": 3.1818181818181817,
91
+ "grad_norm": 1.6458451747894287,
92
+ "learning_rate": 4.6941974739181395e-05,
93
+ "loss": 0.1208,
94
+ "step": 70
95
+ },
96
+ {
97
+ "epoch": 3.6363636363636362,
98
+ "grad_norm": 1.5783404111862183,
99
+ "learning_rate": 4.6031338320779534e-05,
100
+ "loss": 0.1211,
101
+ "step": 80
102
+ },
103
+ {
104
+ "epoch": 4.0,
105
+ "eval_accuracy": 0.8392857142857143,
106
+ "eval_f1": 0.8301314459049545,
107
+ "eval_loss": 0.1286560595035553,
108
+ "eval_precision": 0.8238709677419356,
109
+ "eval_recall": 0.8669590643274854,
110
+ "eval_runtime": 1.9937,
111
+ "eval_samples_per_second": 28.088,
112
+ "eval_steps_per_second": 2.006,
113
+ "step": 88
114
+ },
115
+ {
116
+ "epoch": 4.090909090909091,
117
+ "grad_norm": 2.3769476413726807,
118
+ "learning_rate": 4.5013531023109014e-05,
119
+ "loss": 0.1136,
120
+ "step": 90
121
+ },
122
+ {
123
+ "epoch": 4.545454545454545,
124
+ "grad_norm": 2.6184587478637695,
125
+ "learning_rate": 4.389373935885646e-05,
126
+ "loss": 0.1024,
127
+ "step": 100
128
+ },
129
+ {
130
+ "epoch": 5.0,
131
+ "grad_norm": 2.965785264968872,
132
+ "learning_rate": 4.267766952966369e-05,
133
+ "loss": 0.1159,
134
+ "step": 110
135
+ },
136
+ {
137
+ "epoch": 5.0,
138
+ "eval_accuracy": 0.8392857142857143,
139
+ "eval_f1": 0.790610718737017,
140
+ "eval_loss": 0.08838049322366714,
141
+ "eval_precision": 0.8656565656565656,
142
+ "eval_recall": 0.7646198830409356,
143
+ "eval_runtime": 2.0259,
144
+ "eval_samples_per_second": 27.642,
145
+ "eval_steps_per_second": 1.974,
146
+ "step": 110
147
+ },
148
+ {
149
+ "epoch": 5.454545454545454,
150
+ "grad_norm": 6.49717378616333,
151
+ "learning_rate": 4.137151834863213e-05,
152
+ "loss": 0.1013,
153
+ "step": 120
154
+ },
155
+ {
156
+ "epoch": 5.909090909090909,
157
+ "grad_norm": 2.3494088649749756,
158
+ "learning_rate": 3.9981941662783674e-05,
159
+ "loss": 0.1307,
160
+ "step": 130
161
+ },
162
+ {
163
+ "epoch": 6.0,
164
+ "eval_accuracy": 0.7857142857142857,
165
+ "eval_f1": 0.6818181818181819,
166
+ "eval_loss": 0.13052888214588165,
167
+ "eval_precision": 0.88,
168
+ "eval_recall": 0.6666666666666666,
169
+ "eval_runtime": 2.0536,
170
+ "eval_samples_per_second": 27.269,
171
+ "eval_steps_per_second": 1.948,
172
+ "step": 132
173
+ },
174
+ {
175
+ "epoch": 6.363636363636363,
176
+ "grad_norm": 1.3006728887557983,
177
+ "learning_rate": 3.851602043638994e-05,
178
+ "loss": 0.0906,
179
+ "step": 140
180
+ },
181
+ {
182
+ "epoch": 6.818181818181818,
183
+ "grad_norm": 2.7580723762512207,
184
+ "learning_rate": 3.6981224668001424e-05,
185
+ "loss": 0.1085,
186
+ "step": 150
187
+ },
188
+ {
189
+ "epoch": 7.0,
190
+ "eval_accuracy": 0.8214285714285714,
191
+ "eval_f1": 0.78125,
192
+ "eval_loss": 0.1015724316239357,
193
+ "eval_precision": 0.8095238095238095,
194
+ "eval_recall": 0.7660818713450293,
195
+ "eval_runtime": 2.0225,
196
+ "eval_samples_per_second": 27.688,
197
+ "eval_steps_per_second": 1.978,
198
+ "step": 154
199
+ },
200
+ {
201
+ "epoch": 7.2727272727272725,
202
+ "grad_norm": 2.479637384414673,
203
+ "learning_rate": 3.5385375325047166e-05,
204
+ "loss": 0.0981,
205
+ "step": 160
206
+ },
207
+ {
208
+ "epoch": 7.7272727272727275,
209
+ "grad_norm": 2.572568655014038,
210
+ "learning_rate": 3.3736604489977466e-05,
211
+ "loss": 0.0782,
212
+ "step": 170
213
+ },
214
+ {
215
+ "epoch": 8.0,
216
+ "eval_accuracy": 0.8571428571428571,
217
+ "eval_f1": 0.8095238095238095,
218
+ "eval_loss": 0.06962920725345612,
219
+ "eval_precision": 0.9130434782608696,
220
+ "eval_recall": 0.7777777777777778,
221
+ "eval_runtime": 2.0794,
222
+ "eval_samples_per_second": 26.931,
223
+ "eval_steps_per_second": 1.924,
224
+ "step": 176
225
+ },
226
+ {
227
+ "epoch": 8.181818181818182,
228
+ "grad_norm": 22.004362106323242,
229
+ "learning_rate": 3.2043313921035743e-05,
230
+ "loss": 0.0619,
231
+ "step": 180
232
+ },
233
+ {
234
+ "epoch": 8.636363636363637,
235
+ "grad_norm": 6.473495006561279,
236
+ "learning_rate": 3.0314132238824415e-05,
237
+ "loss": 0.0449,
238
+ "step": 190
239
+ },
240
+ {
241
+ "epoch": 9.0,
242
+ "eval_accuracy": 0.8571428571428571,
243
+ "eval_f1": 0.8095238095238095,
244
+ "eval_loss": 0.1630028635263443,
245
+ "eval_precision": 0.9130434782608696,
246
+ "eval_recall": 0.7777777777777778,
247
+ "eval_runtime": 2.0328,
248
+ "eval_samples_per_second": 27.548,
249
+ "eval_steps_per_second": 1.968,
250
+ "step": 198
251
+ }
252
+ ],
253
+ "logging_steps": 10,
254
+ "max_steps": 440,
255
+ "num_input_tokens_seen": 0,
256
+ "num_train_epochs": 20,
257
+ "save_steps": 500,
258
+ "stateful_callbacks": {
259
+ "EarlyStoppingCallback": {
260
+ "args": {
261
+ "early_stopping_patience": 5,
262
+ "early_stopping_threshold": 0.0
263
+ },
264
+ "attributes": {
265
+ "early_stopping_patience_counter": 5
266
+ }
267
+ },
268
+ "TrainerControl": {
269
+ "args": {
270
+ "should_epoch_stop": false,
271
+ "should_evaluate": false,
272
+ "should_log": false,
273
+ "should_save": true,
274
+ "should_training_stop": true
275
+ },
276
+ "attributes": {}
277
+ }
278
+ },
279
+ "total_flos": 2.4326295022503936e+17,
280
+ "train_batch_size": 16,
281
+ "trial_name": null,
282
+ "trial_params": null
283
+ }
checkpoint-198/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4131706a1be75ab207c26caadf33b032a66d070f6aa64407bea750b6e5973571
3
+ size 5240