Hugging Face's logo

rossijakob
/
silver-dray

Image Classification
Safetensors
English
swin
computer-vision
object-detection
roadwork
Model card Files
xet
 Community
rossijakob commited on
Commit
85302c9
·
verified ·
1 Parent(s): 07015f3

Upload model + model card

Browse files
Files changed (38) hide show
  1. README.md +24 -0
  2. checkpoint-1408/config.json +61 -0
  3. checkpoint-1408/model.safetensors +3 -0
  4. checkpoint-1408/optimizer.pt +3 -0
  5. checkpoint-1408/preprocessor_config.json +22 -0
  6. checkpoint-1408/rng_state.pth +3 -0
  7. checkpoint-1408/scheduler.pt +3 -0
  8. checkpoint-1408/trainer_state.json +253 -0
  9. checkpoint-1408/training_args.bin +3 -0
  10. checkpoint-2112/config.json +61 -0
  11. checkpoint-2112/model.safetensors +3 -0
  12. checkpoint-2112/optimizer.pt +3 -0
  13. checkpoint-2112/preprocessor_config.json +22 -0
  14. checkpoint-2112/rng_state.pth +3 -0
  15. checkpoint-2112/scheduler.pt +3 -0
  16. checkpoint-2112/trainer_state.json +363 -0
  17. checkpoint-2112/training_args.bin +3 -0
  18. checkpoint-2816/config.json +61 -0
  19. checkpoint-2816/model.safetensors +3 -0
  20. checkpoint-2816/optimizer.pt +3 -0
  21. checkpoint-2816/preprocessor_config.json +22 -0
  22. checkpoint-2816/rng_state.pth +3 -0
  23. checkpoint-2816/scheduler.pt +3 -0
  24. checkpoint-2816/trainer_state.json +473 -0
  25. checkpoint-2816/training_args.bin +3 -0
  26. checkpoint-704/config.json +61 -0
  27. checkpoint-704/model.safetensors +3 -0
  28. checkpoint-704/optimizer.pt +3 -0
  29. checkpoint-704/preprocessor_config.json +22 -0
  30. checkpoint-704/rng_state.pth +3 -0
  31. checkpoint-704/scheduler.pt +3 -0
  32. checkpoint-704/trainer_state.json +143 -0
  33. checkpoint-704/training_args.bin +3 -0
  34. config.json +61 -0
  35. model.safetensors +3 -0
  36. model_card.json +7 -0
  37. preprocessor_config.json +22 -0
  38. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ tags:
4
+ - computer-vision
5
+ - object-detection
6
+ - roadwork
7
+ license: mit
8
+ model_type: swin-transformer
9
+ base_model: microsoft/swin-base-patch4-window7-224
10
+ datasets:
11
+ - custom
12
+ pipeline_tag: image-classification
13
+ ---
14
+
15
+ # silver-ray
16
+
17
+ **Description:** roadwork-detect model
18
+ **Version:** 1.0.0
19
+ **Submitted by:** 5C61mvbduMdLp9qvWjyiLeizjNCKQgCtoKv6g8vuTgxjnd8W
20
+ **Submission time:** 1760220656
21
+
22
+ ---
23
+
24
+ This model is fine-tuned on roadwork images using Swin Transformer for roadwork detection.
checkpoint-1408/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
+ "id2label": {
20
+ "0": "no_roadwork",
21
+ "1": "roadwork"
22
+ },
23
+ "image_size": 224,
24
+ "initializer_range": 0.02,
25
+ "label2id": {
26
+ "no_roadwork": 0,
27
+ "roadwork": 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": "4.44.0",
59
+ "use_absolute_embeddings": false,
60
+ "window_size": 7
61
+ }
checkpoint-1408/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61de383927f2323811c05ab06e21e1c99a6ef9338ac95e066d454cf0996b66fa
3
+ size 347498816
checkpoint-1408/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d7f4fee9f0180d60ed4bc5893727f1c1f04331e0270f18c62f7590a84328b7a
3
+ size 694318110
checkpoint-1408/preprocessor_config.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "do_rescale": true,
4
+ "do_resize": true,
5
+ "image_mean": [
6
+ 0.485,
7
+ 0.456,
8
+ 0.406
9
+ ],
10
+ "image_processor_type": "ViTImageProcessor",
11
+ "image_std": [
12
+ 0.229,
13
+ 0.224,
14
+ 0.225
15
+ ],
16
+ "resample": 3,
17
+ "rescale_factor": 0.00392156862745098,
18
+ "size": {
19
+ "height": 224,
20
+ "width": 224
21
+ }
22
+ }
checkpoint-1408/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aea4d7303e6f6a99ad96597ff51522a7f9cb3e1c3a75652d6bce4f0ebdf79953
3
+ size 14645
checkpoint-1408/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ceb005bb446f171dcdf9815b7213fd8a1a8f40a8acb0b5e9d2e207180372a317
3
+ size 1465
checkpoint-1408/trainer_state.json ADDED
@@ -0,0 +1,253 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.962536023054755,
3
+ "best_model_checkpoint": "./roadwork-swin-finetuned/checkpoint-704",
4
+ "epoch": 2.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1408,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.07102272727272728,
13
+ "grad_norm": 17.786890029907227,
14
+ "learning_rate": 2.9488636363636363e-05,
15
+ "loss": 0.416,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.14204545454545456,
20
+ "grad_norm": 23.79692268371582,
21
+ "learning_rate": 2.895596590909091e-05,
22
+ "loss": 0.2916,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.21306818181818182,
27
+ "grad_norm": 19.02972412109375,
28
+ "learning_rate": 2.8433948863636362e-05,
29
+ "loss": 0.2737,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.2840909090909091,
34
+ "grad_norm": 0.4028087556362152,
35
+ "learning_rate": 2.790127840909091e-05,
36
+ "loss": 0.375,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.35511363636363635,
41
+ "grad_norm": 11.175421714782715,
42
+ "learning_rate": 2.7368607954545453e-05,
43
+ "loss": 0.3105,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.42613636363636365,
48
+ "grad_norm": 14.542905807495117,
49
+ "learning_rate": 2.68359375e-05,
50
+ "loss": 0.2578,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.4971590909090909,
55
+ "grad_norm": 0.4432804584503174,
56
+ "learning_rate": 2.630326704545455e-05,
57
+ "loss": 0.3166,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.5681818181818182,
62
+ "grad_norm": 0.9627830386161804,
63
+ "learning_rate": 2.5770596590909092e-05,
64
+ "loss": 0.3289,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.6392045454545454,
69
+ "grad_norm": 5.985296249389648,
70
+ "learning_rate": 2.523792613636364e-05,
71
+ "loss": 0.2113,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.7102272727272727,
76
+ "grad_norm": 6.082040786743164,
77
+ "learning_rate": 2.470525568181818e-05,
78
+ "loss": 0.2488,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.78125,
83
+ "grad_norm": 0.5786997079849243,
84
+ "learning_rate": 2.4172585227272728e-05,
85
+ "loss": 0.2465,
86
+ "step": 550
87
+ },
88
+ {
89
+ "epoch": 0.8522727272727273,
90
+ "grad_norm": 18.639421463012695,
91
+ "learning_rate": 2.3639914772727272e-05,
92
+ "loss": 0.2751,
93
+ "step": 600
94
+ },
95
+ {
96
+ "epoch": 0.9232954545454546,
97
+ "grad_norm": 15.852656364440918,
98
+ "learning_rate": 2.310724431818182e-05,
99
+ "loss": 0.2341,
100
+ "step": 650
101
+ },
102
+ {
103
+ "epoch": 0.9943181818181818,
104
+ "grad_norm": 31.554357528686523,
105
+ "learning_rate": 2.2574573863636364e-05,
106
+ "loss": 0.2625,
107
+ "step": 700
108
+ },
109
+ {
110
+ "epoch": 1.0,
111
+ "eval_accuracy": 0.9376996805111821,
112
+ "eval_f1": 0.962536023054755,
113
+ "eval_loss": 0.24441801011562347,
114
+ "eval_precision": 0.9597701149425287,
115
+ "eval_recall": 0.9653179190751445,
116
+ "eval_runtime": 5.9751,
117
+ "eval_samples_per_second": 104.769,
118
+ "eval_steps_per_second": 13.222,
119
+ "step": 704
120
+ },
121
+ {
122
+ "epoch": 1.0653409090909092,
123
+ "grad_norm": 0.20507046580314636,
124
+ "learning_rate": 2.204190340909091e-05,
125
+ "loss": 0.2122,
126
+ "step": 750
127
+ },
128
+ {
129
+ "epoch": 1.1363636363636362,
130
+ "grad_norm": 7.888465404510498,
131
+ "learning_rate": 2.1509232954545455e-05,
132
+ "loss": 0.1566,
133
+ "step": 800
134
+ },
135
+ {
136
+ "epoch": 1.2073863636363638,
137
+ "grad_norm": 1.0586321353912354,
138
+ "learning_rate": 2.09765625e-05,
139
+ "loss": 0.1512,
140
+ "step": 850
141
+ },
142
+ {
143
+ "epoch": 1.2784090909090908,
144
+ "grad_norm": 8.838766098022461,
145
+ "learning_rate": 2.0443892045454547e-05,
146
+ "loss": 0.2021,
147
+ "step": 900
148
+ },
149
+ {
150
+ "epoch": 1.3494318181818181,
151
+ "grad_norm": 0.0867004320025444,
152
+ "learning_rate": 1.991122159090909e-05,
153
+ "loss": 0.1948,
154
+ "step": 950
155
+ },
156
+ {
157
+ "epoch": 1.4204545454545454,
158
+ "grad_norm": NaN,
159
+ "learning_rate": 1.9399857954545453e-05,
160
+ "loss": 0.2295,
161
+ "step": 1000
162
+ },
163
+ {
164
+ "epoch": 1.4914772727272727,
165
+ "grad_norm": 0.10102058947086334,
166
+ "learning_rate": 1.88671875e-05,
167
+ "loss": 0.1703,
168
+ "step": 1050
169
+ },
170
+ {
171
+ "epoch": 1.5625,
172
+ "grad_norm": 0.09882424771785736,
173
+ "learning_rate": 1.8334517045454548e-05,
174
+ "loss": 0.24,
175
+ "step": 1100
176
+ },
177
+ {
178
+ "epoch": 1.6335227272727273,
179
+ "grad_norm": 2.676793336868286,
180
+ "learning_rate": 1.7801846590909092e-05,
181
+ "loss": 0.1699,
182
+ "step": 1150
183
+ },
184
+ {
185
+ "epoch": 1.7045454545454546,
186
+ "grad_norm": 1.2506362199783325,
187
+ "learning_rate": 1.726917613636364e-05,
188
+ "loss": 0.1621,
189
+ "step": 1200
190
+ },
191
+ {
192
+ "epoch": 1.7755681818181817,
193
+ "grad_norm": 0.06482065469026566,
194
+ "learning_rate": 1.673650568181818e-05,
195
+ "loss": 0.1651,
196
+ "step": 1250
197
+ },
198
+ {
199
+ "epoch": 1.8465909090909092,
200
+ "grad_norm": 0.20499753952026367,
201
+ "learning_rate": 1.6203835227272727e-05,
202
+ "loss": 0.2131,
203
+ "step": 1300
204
+ },
205
+ {
206
+ "epoch": 1.9176136363636362,
207
+ "grad_norm": 41.24250030517578,
208
+ "learning_rate": 1.567116477272727e-05,
209
+ "loss": 0.223,
210
+ "step": 1350
211
+ },
212
+ {
213
+ "epoch": 1.9886363636363638,
214
+ "grad_norm": 131.9430694580078,
215
+ "learning_rate": 1.5138494318181819e-05,
216
+ "loss": 0.132,
217
+ "step": 1400
218
+ },
219
+ {
220
+ "epoch": 2.0,
221
+ "eval_accuracy": 0.9265175718849841,
222
+ "eval_f1": 0.9562737642585551,
223
+ "eval_loss": 0.3970772624015808,
224
+ "eval_precision": 0.9437148217636022,
225
+ "eval_recall": 0.9691714836223507,
226
+ "eval_runtime": 5.9491,
227
+ "eval_samples_per_second": 105.226,
228
+ "eval_steps_per_second": 13.279,
229
+ "step": 1408
230
+ }
231
+ ],
232
+ "logging_steps": 50,
233
+ "max_steps": 2816,
234
+ "num_input_tokens_seen": 0,
235
+ "num_train_epochs": 4,
236
+ "save_steps": 500,
237
+ "stateful_callbacks": {
238
+ "TrainerControl": {
239
+ "args": {
240
+ "should_epoch_stop": false,
241
+ "should_evaluate": false,
242
+ "should_log": false,
243
+ "should_save": true,
244
+ "should_training_stop": false
245
+ },
246
+ "attributes": {}
247
+ }
248
+ },
249
+ "total_flos": 8.8138750081536e+17,
250
+ "train_batch_size": 8,
251
+ "trial_name": null,
252
+ "trial_params": null
253
+ }
checkpoint-1408/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b8a1790f345edd63357c8f19e5c74c8b986b655ff940839229924718dde9f97
3
+ size 5585
checkpoint-2112/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
+ "id2label": {
20
+ "0": "no_roadwork",
21
+ "1": "roadwork"
22
+ },
23
+ "image_size": 224,
24
+ "initializer_range": 0.02,
25
+ "label2id": {
26
+ "no_roadwork": 0,
27
+ "roadwork": 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": "4.44.0",
59
+ "use_absolute_embeddings": false,
60
+ "window_size": 7
61
+ }
checkpoint-2112/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9debc5fe652d7a822ed25dd811114b9bc2bc9b3d799755f838bb94095b10a780
3
+ size 347498816
checkpoint-2112/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e8e7b30a1cf74270fdf6d4daec30cd3fcd401c80d8460b028e8ba8ee43769f5
3
+ size 694318110
checkpoint-2112/preprocessor_config.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "do_rescale": true,
4
+ "do_resize": true,
5
+ "image_mean": [
6
+ 0.485,
7
+ 0.456,
8
+ 0.406
9
+ ],
10
+ "image_processor_type": "ViTImageProcessor",
11
+ "image_std": [
12
+ 0.229,
13
+ 0.224,
14
+ 0.225
15
+ ],
16
+ "resample": 3,
17
+ "rescale_factor": 0.00392156862745098,
18
+ "size": {
19
+ "height": 224,
20
+ "width": 224
21
+ }
22
+ }
checkpoint-2112/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe50a42dbb561b68d1a8f886a19c67ffb7e441f7611a209f5a7b056992934076
3
+ size 14645
checkpoint-2112/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:071554f49f318f3698e276ee34395c7bb452472022fa20c304bc717b37883bf5
3
+ size 1465
checkpoint-2112/trainer_state.json ADDED
@@ -0,0 +1,363 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.962536023054755,
3
+ "best_model_checkpoint": "./roadwork-swin-finetuned/checkpoint-704",
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 2112,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.07102272727272728,
13
+ "grad_norm": 17.786890029907227,
14
+ "learning_rate": 2.9488636363636363e-05,
15
+ "loss": 0.416,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.14204545454545456,
20
+ "grad_norm": 23.79692268371582,
21
+ "learning_rate": 2.895596590909091e-05,
22
+ "loss": 0.2916,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.21306818181818182,
27
+ "grad_norm": 19.02972412109375,
28
+ "learning_rate": 2.8433948863636362e-05,
29
+ "loss": 0.2737,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.2840909090909091,
34
+ "grad_norm": 0.4028087556362152,
35
+ "learning_rate": 2.790127840909091e-05,
36
+ "loss": 0.375,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.35511363636363635,
41
+ "grad_norm": 11.175421714782715,
42
+ "learning_rate": 2.7368607954545453e-05,
43
+ "loss": 0.3105,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.42613636363636365,
48
+ "grad_norm": 14.542905807495117,
49
+ "learning_rate": 2.68359375e-05,
50
+ "loss": 0.2578,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.4971590909090909,
55
+ "grad_norm": 0.4432804584503174,
56
+ "learning_rate": 2.630326704545455e-05,
57
+ "loss": 0.3166,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.5681818181818182,
62
+ "grad_norm": 0.9627830386161804,
63
+ "learning_rate": 2.5770596590909092e-05,
64
+ "loss": 0.3289,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.6392045454545454,
69
+ "grad_norm": 5.985296249389648,
70
+ "learning_rate": 2.523792613636364e-05,
71
+ "loss": 0.2113,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.7102272727272727,
76
+ "grad_norm": 6.082040786743164,
77
+ "learning_rate": 2.470525568181818e-05,
78
+ "loss": 0.2488,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.78125,
83
+ "grad_norm": 0.5786997079849243,
84
+ "learning_rate": 2.4172585227272728e-05,
85
+ "loss": 0.2465,
86
+ "step": 550
87
+ },
88
+ {
89
+ "epoch": 0.8522727272727273,
90
+ "grad_norm": 18.639421463012695,
91
+ "learning_rate": 2.3639914772727272e-05,
92
+ "loss": 0.2751,
93
+ "step": 600
94
+ },
95
+ {
96
+ "epoch": 0.9232954545454546,
97
+ "grad_norm": 15.852656364440918,
98
+ "learning_rate": 2.310724431818182e-05,
99
+ "loss": 0.2341,
100
+ "step": 650
101
+ },
102
+ {
103
+ "epoch": 0.9943181818181818,
104
+ "grad_norm": 31.554357528686523,
105
+ "learning_rate": 2.2574573863636364e-05,
106
+ "loss": 0.2625,
107
+ "step": 700
108
+ },
109
+ {
110
+ "epoch": 1.0,
111
+ "eval_accuracy": 0.9376996805111821,
112
+ "eval_f1": 0.962536023054755,
113
+ "eval_loss": 0.24441801011562347,
114
+ "eval_precision": 0.9597701149425287,
115
+ "eval_recall": 0.9653179190751445,
116
+ "eval_runtime": 5.9751,
117
+ "eval_samples_per_second": 104.769,
118
+ "eval_steps_per_second": 13.222,
119
+ "step": 704
120
+ },
121
+ {
122
+ "epoch": 1.0653409090909092,
123
+ "grad_norm": 0.20507046580314636,
124
+ "learning_rate": 2.204190340909091e-05,
125
+ "loss": 0.2122,
126
+ "step": 750
127
+ },
128
+ {
129
+ "epoch": 1.1363636363636362,
130
+ "grad_norm": 7.888465404510498,
131
+ "learning_rate": 2.1509232954545455e-05,
132
+ "loss": 0.1566,
133
+ "step": 800
134
+ },
135
+ {
136
+ "epoch": 1.2073863636363638,
137
+ "grad_norm": 1.0586321353912354,
138
+ "learning_rate": 2.09765625e-05,
139
+ "loss": 0.1512,
140
+ "step": 850
141
+ },
142
+ {
143
+ "epoch": 1.2784090909090908,
144
+ "grad_norm": 8.838766098022461,
145
+ "learning_rate": 2.0443892045454547e-05,
146
+ "loss": 0.2021,
147
+ "step": 900
148
+ },
149
+ {
150
+ "epoch": 1.3494318181818181,
151
+ "grad_norm": 0.0867004320025444,
152
+ "learning_rate": 1.991122159090909e-05,
153
+ "loss": 0.1948,
154
+ "step": 950
155
+ },
156
+ {
157
+ "epoch": 1.4204545454545454,
158
+ "grad_norm": NaN,
159
+ "learning_rate": 1.9399857954545453e-05,
160
+ "loss": 0.2295,
161
+ "step": 1000
162
+ },
163
+ {
164
+ "epoch": 1.4914772727272727,
165
+ "grad_norm": 0.10102058947086334,
166
+ "learning_rate": 1.88671875e-05,
167
+ "loss": 0.1703,
168
+ "step": 1050
169
+ },
170
+ {
171
+ "epoch": 1.5625,
172
+ "grad_norm": 0.09882424771785736,
173
+ "learning_rate": 1.8334517045454548e-05,
174
+ "loss": 0.24,
175
+ "step": 1100
176
+ },
177
+ {
178
+ "epoch": 1.6335227272727273,
179
+ "grad_norm": 2.676793336868286,
180
+ "learning_rate": 1.7801846590909092e-05,
181
+ "loss": 0.1699,
182
+ "step": 1150
183
+ },
184
+ {
185
+ "epoch": 1.7045454545454546,
186
+ "grad_norm": 1.2506362199783325,
187
+ "learning_rate": 1.726917613636364e-05,
188
+ "loss": 0.1621,
189
+ "step": 1200
190
+ },
191
+ {
192
+ "epoch": 1.7755681818181817,
193
+ "grad_norm": 0.06482065469026566,
194
+ "learning_rate": 1.673650568181818e-05,
195
+ "loss": 0.1651,
196
+ "step": 1250
197
+ },
198
+ {
199
+ "epoch": 1.8465909090909092,
200
+ "grad_norm": 0.20499753952026367,
201
+ "learning_rate": 1.6203835227272727e-05,
202
+ "loss": 0.2131,
203
+ "step": 1300
204
+ },
205
+ {
206
+ "epoch": 1.9176136363636362,
207
+ "grad_norm": 41.24250030517578,
208
+ "learning_rate": 1.567116477272727e-05,
209
+ "loss": 0.223,
210
+ "step": 1350
211
+ },
212
+ {
213
+ "epoch": 1.9886363636363638,
214
+ "grad_norm": 131.9430694580078,
215
+ "learning_rate": 1.5138494318181819e-05,
216
+ "loss": 0.132,
217
+ "step": 1400
218
+ },
219
+ {
220
+ "epoch": 2.0,
221
+ "eval_accuracy": 0.9265175718849841,
222
+ "eval_f1": 0.9562737642585551,
223
+ "eval_loss": 0.3970772624015808,
224
+ "eval_precision": 0.9437148217636022,
225
+ "eval_recall": 0.9691714836223507,
226
+ "eval_runtime": 5.9491,
227
+ "eval_samples_per_second": 105.226,
228
+ "eval_steps_per_second": 13.279,
229
+ "step": 1408
230
+ },
231
+ {
232
+ "epoch": 2.059659090909091,
233
+ "grad_norm": 30.692094802856445,
234
+ "learning_rate": 1.4605823863636365e-05,
235
+ "loss": 0.1146,
236
+ "step": 1450
237
+ },
238
+ {
239
+ "epoch": 2.1306818181818183,
240
+ "grad_norm": 9.568278312683105,
241
+ "learning_rate": 1.407315340909091e-05,
242
+ "loss": 0.0766,
243
+ "step": 1500
244
+ },
245
+ {
246
+ "epoch": 2.2017045454545454,
247
+ "grad_norm": 0.030713092535734177,
248
+ "learning_rate": 1.3540482954545454e-05,
249
+ "loss": 0.1294,
250
+ "step": 1550
251
+ },
252
+ {
253
+ "epoch": 2.2727272727272725,
254
+ "grad_norm": 0.1593484729528427,
255
+ "learning_rate": 1.30078125e-05,
256
+ "loss": 0.1204,
257
+ "step": 1600
258
+ },
259
+ {
260
+ "epoch": 2.34375,
261
+ "grad_norm": 0.012557649984955788,
262
+ "learning_rate": 1.2475142045454546e-05,
263
+ "loss": 0.051,
264
+ "step": 1650
265
+ },
266
+ {
267
+ "epoch": 2.4147727272727275,
268
+ "grad_norm": 64.85875701904297,
269
+ "learning_rate": 1.1942471590909092e-05,
270
+ "loss": 0.178,
271
+ "step": 1700
272
+ },
273
+ {
274
+ "epoch": 2.4857954545454546,
275
+ "grad_norm": 0.0735430121421814,
276
+ "learning_rate": 1.1409801136363636e-05,
277
+ "loss": 0.1308,
278
+ "step": 1750
279
+ },
280
+ {
281
+ "epoch": 2.5568181818181817,
282
+ "grad_norm": 0.07014719396829605,
283
+ "learning_rate": 1.0877130681818182e-05,
284
+ "loss": 0.0831,
285
+ "step": 1800
286
+ },
287
+ {
288
+ "epoch": 2.627840909090909,
289
+ "grad_norm": 19.661163330078125,
290
+ "learning_rate": 1.0344460227272727e-05,
291
+ "loss": 0.1815,
292
+ "step": 1850
293
+ },
294
+ {
295
+ "epoch": 2.6988636363636362,
296
+ "grad_norm": 6.710304260253906,
297
+ "learning_rate": 9.811789772727273e-06,
298
+ "loss": 0.0773,
299
+ "step": 1900
300
+ },
301
+ {
302
+ "epoch": 2.7698863636363638,
303
+ "grad_norm": 0.015027725137770176,
304
+ "learning_rate": 9.279119318181819e-06,
305
+ "loss": 0.0798,
306
+ "step": 1950
307
+ },
308
+ {
309
+ "epoch": 2.840909090909091,
310
+ "grad_norm": 0.021411675959825516,
311
+ "learning_rate": 8.746448863636365e-06,
312
+ "loss": 0.1381,
313
+ "step": 2000
314
+ },
315
+ {
316
+ "epoch": 2.9119318181818183,
317
+ "grad_norm": 36.07814025878906,
318
+ "learning_rate": 8.21377840909091e-06,
319
+ "loss": 0.0631,
320
+ "step": 2050
321
+ },
322
+ {
323
+ "epoch": 2.9829545454545454,
324
+ "grad_norm": 0.362332284450531,
325
+ "learning_rate": 7.681107954545454e-06,
326
+ "loss": 0.1094,
327
+ "step": 2100
328
+ },
329
+ {
330
+ "epoch": 3.0,
331
+ "eval_accuracy": 0.9361022364217252,
332
+ "eval_f1": 0.9614643545279383,
333
+ "eval_loss": 0.34437721967697144,
334
+ "eval_precision": 0.9614643545279383,
335
+ "eval_recall": 0.9614643545279383,
336
+ "eval_runtime": 6.402,
337
+ "eval_samples_per_second": 97.781,
338
+ "eval_steps_per_second": 12.34,
339
+ "step": 2112
340
+ }
341
+ ],
342
+ "logging_steps": 50,
343
+ "max_steps": 2816,
344
+ "num_input_tokens_seen": 0,
345
+ "num_train_epochs": 4,
346
+ "save_steps": 500,
347
+ "stateful_callbacks": {
348
+ "TrainerControl": {
349
+ "args": {
350
+ "should_epoch_stop": false,
351
+ "should_evaluate": false,
352
+ "should_log": false,
353
+ "should_save": true,
354
+ "should_training_stop": false
355
+ },
356
+ "attributes": {}
357
+ }
358
+ },
359
+ "total_flos": 1.32208125122304e+18,
360
+ "train_batch_size": 8,
361
+ "trial_name": null,
362
+ "trial_params": null
363
+ }
checkpoint-2112/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b8a1790f345edd63357c8f19e5c74c8b986b655ff940839229924718dde9f97
3
+ size 5585
checkpoint-2816/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
+ "id2label": {
20
+ "0": "no_roadwork",
21
+ "1": "roadwork"
22
+ },
23
+ "image_size": 224,
24
+ "initializer_range": 0.02,
25
+ "label2id": {
26
+ "no_roadwork": 0,
27
+ "roadwork": 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": "4.44.0",
59
+ "use_absolute_embeddings": false,
60
+ "window_size": 7
61
+ }
checkpoint-2816/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f1c1f8236fa3dd25a190e2c9ff6866bafc246ecd473d18a80a2e82eb3b2b1529
3
+ size 347498816
checkpoint-2816/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e1a5985599a4980065eff6ee3938a1a538ea2aedccc204ed1008cb2472095b9
3
+ size 694318110
checkpoint-2816/preprocessor_config.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "do_rescale": true,
4
+ "do_resize": true,
5
+ "image_mean": [
6
+ 0.485,
7
+ 0.456,
8
+ 0.406
9
+ ],
10
+ "image_processor_type": "ViTImageProcessor",
11
+ "image_std": [
12
+ 0.229,
13
+ 0.224,
14
+ 0.225
15
+ ],
16
+ "resample": 3,
17
+ "rescale_factor": 0.00392156862745098,
18
+ "size": {
19
+ "height": 224,
20
+ "width": 224
21
+ }
22
+ }
checkpoint-2816/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1188097b6f783c5e9e955514497b028cdab565e73cded793e926fdfa4d60d4e5
3
+ size 14645
checkpoint-2816/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd9994f9a6c88eaa858f9956a5b074d4ffa78d4081a4db3295f9766387ecc4d1
3
+ size 1465
checkpoint-2816/trainer_state.json ADDED
@@ -0,0 +1,473 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.962536023054755,
3
+ "best_model_checkpoint": "./roadwork-swin-finetuned/checkpoint-704",
4
+ "epoch": 4.0,
5
+ "eval_steps": 500,
6
+ "global_step": 2816,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.07102272727272728,
13
+ "grad_norm": 17.786890029907227,
14
+ "learning_rate": 2.9488636363636363e-05,
15
+ "loss": 0.416,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.14204545454545456,
20
+ "grad_norm": 23.79692268371582,
21
+ "learning_rate": 2.895596590909091e-05,
22
+ "loss": 0.2916,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.21306818181818182,
27
+ "grad_norm": 19.02972412109375,
28
+ "learning_rate": 2.8433948863636362e-05,
29
+ "loss": 0.2737,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.2840909090909091,
34
+ "grad_norm": 0.4028087556362152,
35
+ "learning_rate": 2.790127840909091e-05,
36
+ "loss": 0.375,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.35511363636363635,
41
+ "grad_norm": 11.175421714782715,
42
+ "learning_rate": 2.7368607954545453e-05,
43
+ "loss": 0.3105,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.42613636363636365,
48
+ "grad_norm": 14.542905807495117,
49
+ "learning_rate": 2.68359375e-05,
50
+ "loss": 0.2578,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.4971590909090909,
55
+ "grad_norm": 0.4432804584503174,
56
+ "learning_rate": 2.630326704545455e-05,
57
+ "loss": 0.3166,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.5681818181818182,
62
+ "grad_norm": 0.9627830386161804,
63
+ "learning_rate": 2.5770596590909092e-05,
64
+ "loss": 0.3289,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.6392045454545454,
69
+ "grad_norm": 5.985296249389648,
70
+ "learning_rate": 2.523792613636364e-05,
71
+ "loss": 0.2113,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.7102272727272727,
76
+ "grad_norm": 6.082040786743164,
77
+ "learning_rate": 2.470525568181818e-05,
78
+ "loss": 0.2488,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.78125,
83
+ "grad_norm": 0.5786997079849243,
84
+ "learning_rate": 2.4172585227272728e-05,
85
+ "loss": 0.2465,
86
+ "step": 550
87
+ },
88
+ {
89
+ "epoch": 0.8522727272727273,
90
+ "grad_norm": 18.639421463012695,
91
+ "learning_rate": 2.3639914772727272e-05,
92
+ "loss": 0.2751,
93
+ "step": 600
94
+ },
95
+ {
96
+ "epoch": 0.9232954545454546,
97
+ "grad_norm": 15.852656364440918,
98
+ "learning_rate": 2.310724431818182e-05,
99
+ "loss": 0.2341,
100
+ "step": 650
101
+ },
102
+ {
103
+ "epoch": 0.9943181818181818,
104
+ "grad_norm": 31.554357528686523,
105
+ "learning_rate": 2.2574573863636364e-05,
106
+ "loss": 0.2625,
107
+ "step": 700
108
+ },
109
+ {
110
+ "epoch": 1.0,
111
+ "eval_accuracy": 0.9376996805111821,
112
+ "eval_f1": 0.962536023054755,
113
+ "eval_loss": 0.24441801011562347,
114
+ "eval_precision": 0.9597701149425287,
115
+ "eval_recall": 0.9653179190751445,
116
+ "eval_runtime": 5.9751,
117
+ "eval_samples_per_second": 104.769,
118
+ "eval_steps_per_second": 13.222,
119
+ "step": 704
120
+ },
121
+ {
122
+ "epoch": 1.0653409090909092,
123
+ "grad_norm": 0.20507046580314636,
124
+ "learning_rate": 2.204190340909091e-05,
125
+ "loss": 0.2122,
126
+ "step": 750
127
+ },
128
+ {
129
+ "epoch": 1.1363636363636362,
130
+ "grad_norm": 7.888465404510498,
131
+ "learning_rate": 2.1509232954545455e-05,
132
+ "loss": 0.1566,
133
+ "step": 800
134
+ },
135
+ {
136
+ "epoch": 1.2073863636363638,
137
+ "grad_norm": 1.0586321353912354,
138
+ "learning_rate": 2.09765625e-05,
139
+ "loss": 0.1512,
140
+ "step": 850
141
+ },
142
+ {
143
+ "epoch": 1.2784090909090908,
144
+ "grad_norm": 8.838766098022461,
145
+ "learning_rate": 2.0443892045454547e-05,
146
+ "loss": 0.2021,
147
+ "step": 900
148
+ },
149
+ {
150
+ "epoch": 1.3494318181818181,
151
+ "grad_norm": 0.0867004320025444,
152
+ "learning_rate": 1.991122159090909e-05,
153
+ "loss": 0.1948,
154
+ "step": 950
155
+ },
156
+ {
157
+ "epoch": 1.4204545454545454,
158
+ "grad_norm": NaN,
159
+ "learning_rate": 1.9399857954545453e-05,
160
+ "loss": 0.2295,
161
+ "step": 1000
162
+ },
163
+ {
164
+ "epoch": 1.4914772727272727,
165
+ "grad_norm": 0.10102058947086334,
166
+ "learning_rate": 1.88671875e-05,
167
+ "loss": 0.1703,
168
+ "step": 1050
169
+ },
170
+ {
171
+ "epoch": 1.5625,
172
+ "grad_norm": 0.09882424771785736,
173
+ "learning_rate": 1.8334517045454548e-05,
174
+ "loss": 0.24,
175
+ "step": 1100
176
+ },
177
+ {
178
+ "epoch": 1.6335227272727273,
179
+ "grad_norm": 2.676793336868286,
180
+ "learning_rate": 1.7801846590909092e-05,
181
+ "loss": 0.1699,
182
+ "step": 1150
183
+ },
184
+ {
185
+ "epoch": 1.7045454545454546,
186
+ "grad_norm": 1.2506362199783325,
187
+ "learning_rate": 1.726917613636364e-05,
188
+ "loss": 0.1621,
189
+ "step": 1200
190
+ },
191
+ {
192
+ "epoch": 1.7755681818181817,
193
+ "grad_norm": 0.06482065469026566,
194
+ "learning_rate": 1.673650568181818e-05,
195
+ "loss": 0.1651,
196
+ "step": 1250
197
+ },
198
+ {
199
+ "epoch": 1.8465909090909092,
200
+ "grad_norm": 0.20499753952026367,
201
+ "learning_rate": 1.6203835227272727e-05,
202
+ "loss": 0.2131,
203
+ "step": 1300
204
+ },
205
+ {
206
+ "epoch": 1.9176136363636362,
207
+ "grad_norm": 41.24250030517578,
208
+ "learning_rate": 1.567116477272727e-05,
209
+ "loss": 0.223,
210
+ "step": 1350
211
+ },
212
+ {
213
+ "epoch": 1.9886363636363638,
214
+ "grad_norm": 131.9430694580078,
215
+ "learning_rate": 1.5138494318181819e-05,
216
+ "loss": 0.132,
217
+ "step": 1400
218
+ },
219
+ {
220
+ "epoch": 2.0,
221
+ "eval_accuracy": 0.9265175718849841,
222
+ "eval_f1": 0.9562737642585551,
223
+ "eval_loss": 0.3970772624015808,
224
+ "eval_precision": 0.9437148217636022,
225
+ "eval_recall": 0.9691714836223507,
226
+ "eval_runtime": 5.9491,
227
+ "eval_samples_per_second": 105.226,
228
+ "eval_steps_per_second": 13.279,
229
+ "step": 1408
230
+ },
231
+ {
232
+ "epoch": 2.059659090909091,
233
+ "grad_norm": 30.692094802856445,
234
+ "learning_rate": 1.4605823863636365e-05,
235
+ "loss": 0.1146,
236
+ "step": 1450
237
+ },
238
+ {
239
+ "epoch": 2.1306818181818183,
240
+ "grad_norm": 9.568278312683105,
241
+ "learning_rate": 1.407315340909091e-05,
242
+ "loss": 0.0766,
243
+ "step": 1500
244
+ },
245
+ {
246
+ "epoch": 2.2017045454545454,
247
+ "grad_norm": 0.030713092535734177,
248
+ "learning_rate": 1.3540482954545454e-05,
249
+ "loss": 0.1294,
250
+ "step": 1550
251
+ },
252
+ {
253
+ "epoch": 2.2727272727272725,
254
+ "grad_norm": 0.1593484729528427,
255
+ "learning_rate": 1.30078125e-05,
256
+ "loss": 0.1204,
257
+ "step": 1600
258
+ },
259
+ {
260
+ "epoch": 2.34375,
261
+ "grad_norm": 0.012557649984955788,
262
+ "learning_rate": 1.2475142045454546e-05,
263
+ "loss": 0.051,
264
+ "step": 1650
265
+ },
266
+ {
267
+ "epoch": 2.4147727272727275,
268
+ "grad_norm": 64.85875701904297,
269
+ "learning_rate": 1.1942471590909092e-05,
270
+ "loss": 0.178,
271
+ "step": 1700
272
+ },
273
+ {
274
+ "epoch": 2.4857954545454546,
275
+ "grad_norm": 0.0735430121421814,
276
+ "learning_rate": 1.1409801136363636e-05,
277
+ "loss": 0.1308,
278
+ "step": 1750
279
+ },
280
+ {
281
+ "epoch": 2.5568181818181817,
282
+ "grad_norm": 0.07014719396829605,
283
+ "learning_rate": 1.0877130681818182e-05,
284
+ "loss": 0.0831,
285
+ "step": 1800
286
+ },
287
+ {
288
+ "epoch": 2.627840909090909,
289
+ "grad_norm": 19.661163330078125,
290
+ "learning_rate": 1.0344460227272727e-05,
291
+ "loss": 0.1815,
292
+ "step": 1850
293
+ },
294
+ {
295
+ "epoch": 2.6988636363636362,
296
+ "grad_norm": 6.710304260253906,
297
+ "learning_rate": 9.811789772727273e-06,
298
+ "loss": 0.0773,
299
+ "step": 1900
300
+ },
301
+ {
302
+ "epoch": 2.7698863636363638,
303
+ "grad_norm": 0.015027725137770176,
304
+ "learning_rate": 9.279119318181819e-06,
305
+ "loss": 0.0798,
306
+ "step": 1950
307
+ },
308
+ {
309
+ "epoch": 2.840909090909091,
310
+ "grad_norm": 0.021411675959825516,
311
+ "learning_rate": 8.746448863636365e-06,
312
+ "loss": 0.1381,
313
+ "step": 2000
314
+ },
315
+ {
316
+ "epoch": 2.9119318181818183,
317
+ "grad_norm": 36.07814025878906,
318
+ "learning_rate": 8.21377840909091e-06,
319
+ "loss": 0.0631,
320
+ "step": 2050
321
+ },
322
+ {
323
+ "epoch": 2.9829545454545454,
324
+ "grad_norm": 0.362332284450531,
325
+ "learning_rate": 7.681107954545454e-06,
326
+ "loss": 0.1094,
327
+ "step": 2100
328
+ },
329
+ {
330
+ "epoch": 3.0,
331
+ "eval_accuracy": 0.9361022364217252,
332
+ "eval_f1": 0.9614643545279383,
333
+ "eval_loss": 0.34437721967697144,
334
+ "eval_precision": 0.9614643545279383,
335
+ "eval_recall": 0.9614643545279383,
336
+ "eval_runtime": 6.402,
337
+ "eval_samples_per_second": 97.781,
338
+ "eval_steps_per_second": 12.34,
339
+ "step": 2112
340
+ },
341
+ {
342
+ "epoch": 3.053977272727273,
343
+ "grad_norm": 1.0269795656204224,
344
+ "learning_rate": 7.1484375e-06,
345
+ "loss": 0.0431,
346
+ "step": 2150
347
+ },
348
+ {
349
+ "epoch": 3.125,
350
+ "grad_norm": 0.06955037266016006,
351
+ "learning_rate": 6.615767045454546e-06,
352
+ "loss": 0.044,
353
+ "step": 2200
354
+ },
355
+ {
356
+ "epoch": 3.196022727272727,
357
+ "grad_norm": 0.2144497185945511,
358
+ "learning_rate": 6.083096590909091e-06,
359
+ "loss": 0.078,
360
+ "step": 2250
361
+ },
362
+ {
363
+ "epoch": 3.2670454545454546,
364
+ "grad_norm": 0.8404433131217957,
365
+ "learning_rate": 5.5504261363636365e-06,
366
+ "loss": 0.0634,
367
+ "step": 2300
368
+ },
369
+ {
370
+ "epoch": 3.3380681818181817,
371
+ "grad_norm": 0.018121657893061638,
372
+ "learning_rate": 5.017755681818182e-06,
373
+ "loss": 0.0263,
374
+ "step": 2350
375
+ },
376
+ {
377
+ "epoch": 3.409090909090909,
378
+ "grad_norm": 0.9491384625434875,
379
+ "learning_rate": 4.485085227272727e-06,
380
+ "loss": 0.0913,
381
+ "step": 2400
382
+ },
383
+ {
384
+ "epoch": 3.4801136363636362,
385
+ "grad_norm": 0.11287333071231842,
386
+ "learning_rate": 3.952414772727273e-06,
387
+ "loss": 0.0747,
388
+ "step": 2450
389
+ },
390
+ {
391
+ "epoch": 3.5511363636363638,
392
+ "grad_norm": 10.997075080871582,
393
+ "learning_rate": 3.4197443181818183e-06,
394
+ "loss": 0.0654,
395
+ "step": 2500
396
+ },
397
+ {
398
+ "epoch": 3.622159090909091,
399
+ "grad_norm": 0.04853343218564987,
400
+ "learning_rate": 2.8870738636363636e-06,
401
+ "loss": 0.0664,
402
+ "step": 2550
403
+ },
404
+ {
405
+ "epoch": 3.6931818181818183,
406
+ "grad_norm": 29.09600067138672,
407
+ "learning_rate": 2.354403409090909e-06,
408
+ "loss": 0.0719,
409
+ "step": 2600
410
+ },
411
+ {
412
+ "epoch": 3.7642045454545454,
413
+ "grad_norm": 0.372402161359787,
414
+ "learning_rate": 1.8217329545454545e-06,
415
+ "loss": 0.0713,
416
+ "step": 2650
417
+ },
418
+ {
419
+ "epoch": 3.8352272727272725,
420
+ "grad_norm": 0.04578320309519768,
421
+ "learning_rate": 1.2890625e-06,
422
+ "loss": 0.0363,
423
+ "step": 2700
424
+ },
425
+ {
426
+ "epoch": 3.90625,
427
+ "grad_norm": 0.3740798234939575,
428
+ "learning_rate": 7.563920454545454e-07,
429
+ "loss": 0.0614,
430
+ "step": 2750
431
+ },
432
+ {
433
+ "epoch": 3.9772727272727275,
434
+ "grad_norm": 0.006558969616889954,
435
+ "learning_rate": 2.2372159090909091e-07,
436
+ "loss": 0.0539,
437
+ "step": 2800
438
+ },
439
+ {
440
+ "epoch": 4.0,
441
+ "eval_accuracy": 0.9329073482428115,
442
+ "eval_f1": 0.9593810444874274,
443
+ "eval_loss": 0.41001200675964355,
444
+ "eval_precision": 0.9631067961165048,
445
+ "eval_recall": 0.9556840077071291,
446
+ "eval_runtime": 5.9591,
447
+ "eval_samples_per_second": 105.049,
448
+ "eval_steps_per_second": 13.257,
449
+ "step": 2816
450
+ }
451
+ ],
452
+ "logging_steps": 50,
453
+ "max_steps": 2816,
454
+ "num_input_tokens_seen": 0,
455
+ "num_train_epochs": 4,
456
+ "save_steps": 500,
457
+ "stateful_callbacks": {
458
+ "TrainerControl": {
459
+ "args": {
460
+ "should_epoch_stop": false,
461
+ "should_evaluate": false,
462
+ "should_log": false,
463
+ "should_save": true,
464
+ "should_training_stop": true
465
+ },
466
+ "attributes": {}
467
+ }
468
+ },
469
+ "total_flos": 1.76277500163072e+18,
470
+ "train_batch_size": 8,
471
+ "trial_name": null,
472
+ "trial_params": null
473
+ }
checkpoint-2816/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b8a1790f345edd63357c8f19e5c74c8b986b655ff940839229924718dde9f97
3
+ size 5585
checkpoint-704/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
+ "id2label": {
20
+ "0": "no_roadwork",
21
+ "1": "roadwork"
22
+ },
23
+ "image_size": 224,
24
+ "initializer_range": 0.02,
25
+ "label2id": {
26
+ "no_roadwork": 0,
27
+ "roadwork": 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": "4.44.0",
59
+ "use_absolute_embeddings": false,
60
+ "window_size": 7
61
+ }
checkpoint-704/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9828104797de1f4c1f666739b9958ae18eaff11c1beb177178448a5347d77779
3
+ size 347498816
checkpoint-704/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:918319ac968a008b7bcf60f497f8d88672149faff7e04764d386e5968b7ffde0
3
+ size 694318110
checkpoint-704/preprocessor_config.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "do_rescale": true,
4
+ "do_resize": true,
5
+ "image_mean": [
6
+ 0.485,
7
+ 0.456,
8
+ 0.406
9
+ ],
10
+ "image_processor_type": "ViTImageProcessor",
11
+ "image_std": [
12
+ 0.229,
13
+ 0.224,
14
+ 0.225
15
+ ],
16
+ "resample": 3,
17
+ "rescale_factor": 0.00392156862745098,
18
+ "size": {
19
+ "height": 224,
20
+ "width": 224
21
+ }
22
+ }
checkpoint-704/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31a5c2953af7dc9597ae0ad014b1a6ee2144ca586bd9022714549fa291784069
3
+ size 14645
checkpoint-704/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:75afa0284c54290ff4f03351b03710089d449773102b43b31fc425ad9bb32891
3
+ size 1465
checkpoint-704/trainer_state.json ADDED
@@ -0,0 +1,143 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.962536023054755,
3
+ "best_model_checkpoint": "./roadwork-swin-finetuned/checkpoint-704",
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 704,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.07102272727272728,
13
+ "grad_norm": 17.786890029907227,
14
+ "learning_rate": 2.9488636363636363e-05,
15
+ "loss": 0.416,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.14204545454545456,
20
+ "grad_norm": 23.79692268371582,
21
+ "learning_rate": 2.895596590909091e-05,
22
+ "loss": 0.2916,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.21306818181818182,
27
+ "grad_norm": 19.02972412109375,
28
+ "learning_rate": 2.8433948863636362e-05,
29
+ "loss": 0.2737,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.2840909090909091,
34
+ "grad_norm": 0.4028087556362152,
35
+ "learning_rate": 2.790127840909091e-05,
36
+ "loss": 0.375,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.35511363636363635,
41
+ "grad_norm": 11.175421714782715,
42
+ "learning_rate": 2.7368607954545453e-05,
43
+ "loss": 0.3105,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.42613636363636365,
48
+ "grad_norm": 14.542905807495117,
49
+ "learning_rate": 2.68359375e-05,
50
+ "loss": 0.2578,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.4971590909090909,
55
+ "grad_norm": 0.4432804584503174,
56
+ "learning_rate": 2.630326704545455e-05,
57
+ "loss": 0.3166,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.5681818181818182,
62
+ "grad_norm": 0.9627830386161804,
63
+ "learning_rate": 2.5770596590909092e-05,
64
+ "loss": 0.3289,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.6392045454545454,
69
+ "grad_norm": 5.985296249389648,
70
+ "learning_rate": 2.523792613636364e-05,
71
+ "loss": 0.2113,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.7102272727272727,
76
+ "grad_norm": 6.082040786743164,
77
+ "learning_rate": 2.470525568181818e-05,
78
+ "loss": 0.2488,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.78125,
83
+ "grad_norm": 0.5786997079849243,
84
+ "learning_rate": 2.4172585227272728e-05,
85
+ "loss": 0.2465,
86
+ "step": 550
87
+ },
88
+ {
89
+ "epoch": 0.8522727272727273,
90
+ "grad_norm": 18.639421463012695,
91
+ "learning_rate": 2.3639914772727272e-05,
92
+ "loss": 0.2751,
93
+ "step": 600
94
+ },
95
+ {
96
+ "epoch": 0.9232954545454546,
97
+ "grad_norm": 15.852656364440918,
98
+ "learning_rate": 2.310724431818182e-05,
99
+ "loss": 0.2341,
100
+ "step": 650
101
+ },
102
+ {
103
+ "epoch": 0.9943181818181818,
104
+ "grad_norm": 31.554357528686523,
105
+ "learning_rate": 2.2574573863636364e-05,
106
+ "loss": 0.2625,
107
+ "step": 700
108
+ },
109
+ {
110
+ "epoch": 1.0,
111
+ "eval_accuracy": 0.9376996805111821,
112
+ "eval_f1": 0.962536023054755,
113
+ "eval_loss": 0.24441801011562347,
114
+ "eval_precision": 0.9597701149425287,
115
+ "eval_recall": 0.9653179190751445,
116
+ "eval_runtime": 5.9751,
117
+ "eval_samples_per_second": 104.769,
118
+ "eval_steps_per_second": 13.222,
119
+ "step": 704
120
+ }
121
+ ],
122
+ "logging_steps": 50,
123
+ "max_steps": 2816,
124
+ "num_input_tokens_seen": 0,
125
+ "num_train_epochs": 4,
126
+ "save_steps": 500,
127
+ "stateful_callbacks": {
128
+ "TrainerControl": {
129
+ "args": {
130
+ "should_epoch_stop": false,
131
+ "should_evaluate": false,
132
+ "should_log": false,
133
+ "should_save": true,
134
+ "should_training_stop": false
135
+ },
136
+ "attributes": {}
137
+ }
138
+ },
139
+ "total_flos": 4.4069375040768e+17,
140
+ "train_batch_size": 8,
141
+ "trial_name": null,
142
+ "trial_params": null
143
+ }
checkpoint-704/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b8a1790f345edd63357c8f19e5c74c8b986b655ff940839229924718dde9f97
3
+ size 5585
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
+ "id2label": {
20
+ "0": "no_roadwork",
21
+ "1": "roadwork"
22
+ },
23
+ "image_size": 224,
24
+ "initializer_range": 0.02,
25
+ "label2id": {
26
+ "no_roadwork": 0,
27
+ "roadwork": 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": "4.44.0",
59
+ "use_absolute_embeddings": false,
60
+ "window_size": 7
61
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9828104797de1f4c1f666739b9958ae18eaff11c1beb177178448a5347d77779
3
+ size 347498816
model_card.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_name": "rossijakob/rossi-dray",
3
+ "description": "roadwork-detect model",
4
+ "version": "1.0.0",
5
+ "submitted_by": "5C61mvbduMdLp9qvWjyiLeizjNCKQgCtoKv6g8vuTgxjnd8W",
6
+ "submission_time": 1760220856
7
+ }
preprocessor_config.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "do_rescale": true,
4
+ "do_resize": true,
5
+ "image_mean": [
6
+ 0.485,
7
+ 0.456,
8
+ 0.406
9
+ ],
10
+ "image_processor_type": "ViTImageProcessor",
11
+ "image_std": [
12
+ 0.229,
13
+ 0.224,
14
+ 0.225
15
+ ],
16
+ "resample": 3,
17
+ "rescale_factor": 0.00392156862745098,
18
+ "size": {
19
+ "height": 224,
20
+ "width": 224
21
+ }
22
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b8a1790f345edd63357c8f19e5c74c8b986b655ff940839229924718dde9f97
3
+ size 5585