abhishek HF staff commited on
Commit
74de533
1 Parent(s): d97adba

Upload folder using huggingface_hub

Browse files
README.md ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ ---
3
+ tags:
4
+ - autotrain
5
+ - object-detection
6
+ widget:
7
+ - src: https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg
8
+ example_title: Tiger
9
+ - src: https://huggingface.co/datasets/mishig/sample_images/resolve/main/teapot.jpg
10
+ example_title: Teapot
11
+ - src: https://huggingface.co/datasets/mishig/sample_images/resolve/main/palace.jpg
12
+ example_title: Palace
13
+ datasets:
14
+ - cppe-5
15
+ ---
16
+
17
+ # Model Trained Using AutoTrain
18
+
19
+ - Problem type: Object Detection
20
+
21
+ ## Validation Metrics
22
+ loss: 1.7329339981079102
23
+
24
+ map: 0.1341
25
+
26
+ map_50: 0.2559
27
+
28
+ map_75: 0.1179
29
+
30
+ map_small: 0.0652
31
+
32
+ map_medium: 0.0367
33
+
34
+ map_large: 0.1421
35
+
36
+ mar_1: 0.1278
37
+
38
+ mar_10: 0.2553
39
+
40
+ mar_100: 0.2932
41
+
42
+ mar_small: 0.1689
43
+
44
+ mar_medium: 0.1701
45
+
46
+ mar_large: 0.2122
47
+
48
+ map_Coverall: 0.5042
49
+
50
+ mar_100_Coverall: 0.7222
51
+
52
+ map_Face_Shield: 0.0
53
+
54
+ mar_100_Face_Shield: 0.0
55
+
56
+ map_Gloves: 0.0962
57
+
58
+ mar_100_Gloves: 0.3246
59
+
60
+ map_Goggles: 0.0
61
+
62
+ mar_100_Goggles: 0.0
63
+
64
+ map_Mask: 0.07
65
+
66
+ mar_100_Mask: 0.4192
checkpoint-625/config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/detr-resnet-50",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "relu",
5
+ "architectures": [
6
+ "DetrForObjectDetection"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "auxiliary_loss": false,
10
+ "backbone": "resnet50",
11
+ "backbone_config": null,
12
+ "backbone_kwargs": null,
13
+ "bbox_cost": 5,
14
+ "bbox_loss_coefficient": 5,
15
+ "class_cost": 1,
16
+ "classifier_dropout": 0.0,
17
+ "d_model": 256,
18
+ "decoder_attention_heads": 8,
19
+ "decoder_ffn_dim": 2048,
20
+ "decoder_layerdrop": 0.0,
21
+ "decoder_layers": 6,
22
+ "dice_loss_coefficient": 1,
23
+ "dilation": false,
24
+ "dropout": 0.1,
25
+ "encoder_attention_heads": 8,
26
+ "encoder_ffn_dim": 2048,
27
+ "encoder_layerdrop": 0.0,
28
+ "encoder_layers": 6,
29
+ "eos_coefficient": 0.1,
30
+ "giou_cost": 2,
31
+ "giou_loss_coefficient": 2,
32
+ "id2label": {
33
+ "0": "Coverall",
34
+ "1": "Face_Shield",
35
+ "2": "Gloves",
36
+ "3": "Goggles",
37
+ "4": "Mask"
38
+ },
39
+ "init_std": 0.02,
40
+ "init_xavier_std": 1.0,
41
+ "is_encoder_decoder": true,
42
+ "label2id": {
43
+ "Coverall": 0,
44
+ "Face_Shield": 1,
45
+ "Gloves": 2,
46
+ "Goggles": 3,
47
+ "Mask": 4
48
+ },
49
+ "mask_loss_coefficient": 1,
50
+ "max_position_embeddings": 1024,
51
+ "model_type": "detr",
52
+ "num_channels": 3,
53
+ "num_hidden_layers": 6,
54
+ "num_queries": 100,
55
+ "position_embedding_type": "sine",
56
+ "scale_embedding": false,
57
+ "torch_dtype": "float32",
58
+ "transformers_version": "4.40.1",
59
+ "use_pretrained_backbone": true,
60
+ "use_timm_backbone": true
61
+ }
checkpoint-625/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d39d274c25ce58eeb595dda77b61165ac29c61aa45eecf7efcb0e835f3c06d3
3
+ size 166498936
checkpoint-625/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3cfcfabfc419a31464afc7c5ef88bbfbb4f550e56f0fd19a6a4cb77a23d5144a
3
+ size 330499962
checkpoint-625/preprocessor_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_valid_processor_keys": [
3
+ "images",
4
+ "annotations",
5
+ "return_segmentation_masks",
6
+ "masks_path",
7
+ "do_resize",
8
+ "size",
9
+ "resample",
10
+ "do_rescale",
11
+ "rescale_factor",
12
+ "do_normalize",
13
+ "do_convert_annotations",
14
+ "image_mean",
15
+ "image_std",
16
+ "do_pad",
17
+ "format",
18
+ "return_tensors",
19
+ "data_format",
20
+ "input_data_format"
21
+ ],
22
+ "do_convert_annotations": true,
23
+ "do_normalize": true,
24
+ "do_pad": false,
25
+ "do_rescale": true,
26
+ "do_resize": false,
27
+ "format": "coco_detection",
28
+ "image_mean": [
29
+ 0.485,
30
+ 0.456,
31
+ 0.406
32
+ ],
33
+ "image_processor_type": "DetrImageProcessor",
34
+ "image_std": [
35
+ 0.229,
36
+ 0.224,
37
+ 0.225
38
+ ],
39
+ "resample": 2,
40
+ "rescale_factor": 0.00392156862745098,
41
+ "size": {
42
+ "longest_edge": 600
43
+ }
44
+ }
checkpoint-625/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc76ec9c7860731098d9eaa1906a8cbbf01732b46f7e925f5ef15b27e8038e16
3
+ size 14244
checkpoint-625/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf781e0f0c69c78f4f7dc0bd8dc64143b5d3ec4b2317d8401472d359a60092bd
3
+ size 1064
checkpoint-625/trainer_state.json ADDED
@@ -0,0 +1,605 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.7329339981079102,
3
+ "best_model_checkpoint": "autotrain-obj-det-test-1/checkpoint-625",
4
+ "epoch": 5.0,
5
+ "eval_steps": 500,
6
+ "global_step": 625,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.08,
13
+ "grad_norm": 119.73218536376953,
14
+ "learning_rate": 5.555555555555556e-06,
15
+ "loss": 3.7169,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.16,
20
+ "grad_norm": 618.2245483398438,
21
+ "learning_rate": 1.1904761904761905e-05,
22
+ "loss": 3.9737,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.24,
27
+ "grad_norm": 204.22610473632812,
28
+ "learning_rate": 1.984126984126984e-05,
29
+ "loss": 3.5056,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.32,
34
+ "grad_norm": 177.1145477294922,
35
+ "learning_rate": 2.777777777777778e-05,
36
+ "loss": 3.271,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.4,
41
+ "grad_norm": 163.36953735351562,
42
+ "learning_rate": 3.571428571428572e-05,
43
+ "loss": 3.0842,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.48,
48
+ "grad_norm": 51.06690216064453,
49
+ "learning_rate": 4.3650793650793655e-05,
50
+ "loss": 2.7783,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.56,
55
+ "grad_norm": 77.7472152709961,
56
+ "learning_rate": 4.9822064056939506e-05,
57
+ "loss": 2.7264,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.64,
62
+ "grad_norm": 99.1742935180664,
63
+ "learning_rate": 4.893238434163701e-05,
64
+ "loss": 2.4674,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.72,
69
+ "grad_norm": 77.92932891845703,
70
+ "learning_rate": 4.8042704626334526e-05,
71
+ "loss": 2.5182,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.8,
76
+ "grad_norm": 80.95340728759766,
77
+ "learning_rate": 4.7153024911032026e-05,
78
+ "loss": 2.3434,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.88,
83
+ "grad_norm": 30.60334014892578,
84
+ "learning_rate": 4.635231316725978e-05,
85
+ "loss": 2.5583,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.96,
90
+ "grad_norm": 46.196781158447266,
91
+ "learning_rate": 4.5462633451957297e-05,
92
+ "loss": 2.3346,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 1.0,
97
+ "eval_loss": 2.443927526473999,
98
+ "eval_map": 0.0143,
99
+ "eval_map_50": 0.0359,
100
+ "eval_map_75": 0.0093,
101
+ "eval_map_Coverall": 0.0411,
102
+ "eval_map_Face_Shield": 0.0,
103
+ "eval_map_Gloves": 0.0229,
104
+ "eval_map_Goggles": 0.0,
105
+ "eval_map_Mask": 0.0074,
106
+ "eval_map_large": 0.0347,
107
+ "eval_map_medium": 0.0056,
108
+ "eval_map_small": 0.0398,
109
+ "eval_mar_1": 0.0396,
110
+ "eval_mar_10": 0.1024,
111
+ "eval_mar_100": 0.1618,
112
+ "eval_mar_100_Coverall": 0.4756,
113
+ "eval_mar_100_Face_Shield": 0.0,
114
+ "eval_mar_100_Gloves": 0.0967,
115
+ "eval_mar_100_Goggles": 0.0,
116
+ "eval_mar_100_Mask": 0.2365,
117
+ "eval_mar_large": 0.1741,
118
+ "eval_mar_medium": 0.1149,
119
+ "eval_mar_small": 0.0548,
120
+ "eval_runtime": 0.9557,
121
+ "eval_samples_per_second": 30.345,
122
+ "eval_steps_per_second": 2.093,
123
+ "step": 125
124
+ },
125
+ {
126
+ "epoch": 1.04,
127
+ "grad_norm": 48.32860565185547,
128
+ "learning_rate": 4.45729537366548e-05,
129
+ "loss": 2.4065,
130
+ "step": 130
131
+ },
132
+ {
133
+ "epoch": 1.12,
134
+ "grad_norm": 27.714534759521484,
135
+ "learning_rate": 4.3683274021352316e-05,
136
+ "loss": 2.2728,
137
+ "step": 140
138
+ },
139
+ {
140
+ "epoch": 1.2,
141
+ "grad_norm": 50.94224166870117,
142
+ "learning_rate": 4.279359430604982e-05,
143
+ "loss": 2.0997,
144
+ "step": 150
145
+ },
146
+ {
147
+ "epoch": 1.28,
148
+ "grad_norm": 194.94711303710938,
149
+ "learning_rate": 4.1903914590747336e-05,
150
+ "loss": 2.2642,
151
+ "step": 160
152
+ },
153
+ {
154
+ "epoch": 1.3599999999999999,
155
+ "grad_norm": 86.10822296142578,
156
+ "learning_rate": 4.101423487544484e-05,
157
+ "loss": 2.731,
158
+ "step": 170
159
+ },
160
+ {
161
+ "epoch": 1.44,
162
+ "grad_norm": 107.91861724853516,
163
+ "learning_rate": 4.0124555160142355e-05,
164
+ "loss": 2.3309,
165
+ "step": 180
166
+ },
167
+ {
168
+ "epoch": 1.52,
169
+ "grad_norm": 48.93113708496094,
170
+ "learning_rate": 3.9234875444839855e-05,
171
+ "loss": 2.0455,
172
+ "step": 190
173
+ },
174
+ {
175
+ "epoch": 1.6,
176
+ "grad_norm": 43.76406478881836,
177
+ "learning_rate": 3.834519572953737e-05,
178
+ "loss": 2.0899,
179
+ "step": 200
180
+ },
181
+ {
182
+ "epoch": 1.6800000000000002,
183
+ "grad_norm": 32.99532699584961,
184
+ "learning_rate": 3.7455516014234874e-05,
185
+ "loss": 2.1082,
186
+ "step": 210
187
+ },
188
+ {
189
+ "epoch": 1.76,
190
+ "grad_norm": 25.14192008972168,
191
+ "learning_rate": 3.656583629893239e-05,
192
+ "loss": 2.0704,
193
+ "step": 220
194
+ },
195
+ {
196
+ "epoch": 1.8399999999999999,
197
+ "grad_norm": 80.21981811523438,
198
+ "learning_rate": 3.5676156583629894e-05,
199
+ "loss": 2.0651,
200
+ "step": 230
201
+ },
202
+ {
203
+ "epoch": 1.92,
204
+ "grad_norm": 43.88581085205078,
205
+ "learning_rate": 3.478647686832741e-05,
206
+ "loss": 2.0387,
207
+ "step": 240
208
+ },
209
+ {
210
+ "epoch": 2.0,
211
+ "grad_norm": 44.32549285888672,
212
+ "learning_rate": 3.3896797153024914e-05,
213
+ "loss": 2.2185,
214
+ "step": 250
215
+ },
216
+ {
217
+ "epoch": 2.0,
218
+ "eval_loss": 2.12321400642395,
219
+ "eval_map": 0.1024,
220
+ "eval_map_50": 0.2006,
221
+ "eval_map_75": 0.0995,
222
+ "eval_map_Coverall": 0.4129,
223
+ "eval_map_Face_Shield": 0.0,
224
+ "eval_map_Gloves": 0.0595,
225
+ "eval_map_Goggles": 0.0,
226
+ "eval_map_Mask": 0.0396,
227
+ "eval_map_large": 0.1113,
228
+ "eval_map_medium": 0.0266,
229
+ "eval_map_small": 0.0299,
230
+ "eval_mar_1": 0.1064,
231
+ "eval_mar_10": 0.2095,
232
+ "eval_mar_100": 0.2375,
233
+ "eval_mar_100_Coverall": 0.6667,
234
+ "eval_mar_100_Face_Shield": 0.0,
235
+ "eval_mar_100_Gloves": 0.2459,
236
+ "eval_mar_100_Goggles": 0.0,
237
+ "eval_mar_100_Mask": 0.275,
238
+ "eval_mar_large": 0.1603,
239
+ "eval_mar_medium": 0.1997,
240
+ "eval_mar_small": 0.0973,
241
+ "eval_runtime": 0.9234,
242
+ "eval_samples_per_second": 31.405,
243
+ "eval_steps_per_second": 2.166,
244
+ "step": 250
245
+ },
246
+ {
247
+ "epoch": 2.08,
248
+ "grad_norm": 58.51893997192383,
249
+ "learning_rate": 3.300711743772243e-05,
250
+ "loss": 2.0798,
251
+ "step": 260
252
+ },
253
+ {
254
+ "epoch": 2.16,
255
+ "grad_norm": 79.92374420166016,
256
+ "learning_rate": 3.2117437722419926e-05,
257
+ "loss": 2.0807,
258
+ "step": 270
259
+ },
260
+ {
261
+ "epoch": 2.24,
262
+ "grad_norm": 46.597137451171875,
263
+ "learning_rate": 3.122775800711744e-05,
264
+ "loss": 2.0171,
265
+ "step": 280
266
+ },
267
+ {
268
+ "epoch": 2.32,
269
+ "grad_norm": 32.78828048706055,
270
+ "learning_rate": 3.0338078291814946e-05,
271
+ "loss": 1.8417,
272
+ "step": 290
273
+ },
274
+ {
275
+ "epoch": 2.4,
276
+ "grad_norm": 46.39910125732422,
277
+ "learning_rate": 2.944839857651246e-05,
278
+ "loss": 2.1003,
279
+ "step": 300
280
+ },
281
+ {
282
+ "epoch": 2.48,
283
+ "grad_norm": 38.146358489990234,
284
+ "learning_rate": 2.8558718861209965e-05,
285
+ "loss": 1.8223,
286
+ "step": 310
287
+ },
288
+ {
289
+ "epoch": 2.56,
290
+ "grad_norm": 33.644989013671875,
291
+ "learning_rate": 2.7758007117437723e-05,
292
+ "loss": 1.9298,
293
+ "step": 320
294
+ },
295
+ {
296
+ "epoch": 2.64,
297
+ "grad_norm": 29.090211868286133,
298
+ "learning_rate": 2.6868327402135236e-05,
299
+ "loss": 1.7866,
300
+ "step": 330
301
+ },
302
+ {
303
+ "epoch": 2.7199999999999998,
304
+ "grad_norm": 81.5531005859375,
305
+ "learning_rate": 2.597864768683274e-05,
306
+ "loss": 1.8001,
307
+ "step": 340
308
+ },
309
+ {
310
+ "epoch": 2.8,
311
+ "grad_norm": 35.4053840637207,
312
+ "learning_rate": 2.5088967971530253e-05,
313
+ "loss": 2.0191,
314
+ "step": 350
315
+ },
316
+ {
317
+ "epoch": 2.88,
318
+ "grad_norm": 22.44593620300293,
319
+ "learning_rate": 2.419928825622776e-05,
320
+ "loss": 1.8575,
321
+ "step": 360
322
+ },
323
+ {
324
+ "epoch": 2.96,
325
+ "grad_norm": 549.6776123046875,
326
+ "learning_rate": 2.330960854092527e-05,
327
+ "loss": 2.04,
328
+ "step": 370
329
+ },
330
+ {
331
+ "epoch": 3.0,
332
+ "eval_loss": 2.0039143562316895,
333
+ "eval_map": 0.1056,
334
+ "eval_map_50": 0.2143,
335
+ "eval_map_75": 0.0918,
336
+ "eval_map_Coverall": 0.4461,
337
+ "eval_map_Face_Shield": 0.0,
338
+ "eval_map_Gloves": 0.0531,
339
+ "eval_map_Goggles": 0.0,
340
+ "eval_map_Mask": 0.0288,
341
+ "eval_map_large": 0.0995,
342
+ "eval_map_medium": 0.0161,
343
+ "eval_map_small": 0.0487,
344
+ "eval_mar_1": 0.1172,
345
+ "eval_mar_10": 0.2035,
346
+ "eval_mar_100": 0.2492,
347
+ "eval_mar_100_Coverall": 0.6622,
348
+ "eval_mar_100_Face_Shield": 0.0,
349
+ "eval_mar_100_Gloves": 0.2492,
350
+ "eval_mar_100_Goggles": 0.0,
351
+ "eval_mar_100_Mask": 0.3346,
352
+ "eval_mar_large": 0.2226,
353
+ "eval_mar_medium": 0.1424,
354
+ "eval_mar_small": 0.0988,
355
+ "eval_runtime": 0.9381,
356
+ "eval_samples_per_second": 30.914,
357
+ "eval_steps_per_second": 2.132,
358
+ "step": 375
359
+ },
360
+ {
361
+ "epoch": 3.04,
362
+ "grad_norm": 60.65336990356445,
363
+ "learning_rate": 2.2419928825622775e-05,
364
+ "loss": 1.9005,
365
+ "step": 380
366
+ },
367
+ {
368
+ "epoch": 3.12,
369
+ "grad_norm": 35.07551574707031,
370
+ "learning_rate": 2.1530249110320285e-05,
371
+ "loss": 1.9316,
372
+ "step": 390
373
+ },
374
+ {
375
+ "epoch": 3.2,
376
+ "grad_norm": 25.539474487304688,
377
+ "learning_rate": 2.0640569395017795e-05,
378
+ "loss": 1.901,
379
+ "step": 400
380
+ },
381
+ {
382
+ "epoch": 3.2800000000000002,
383
+ "grad_norm": 38.46770477294922,
384
+ "learning_rate": 1.9750889679715305e-05,
385
+ "loss": 1.6579,
386
+ "step": 410
387
+ },
388
+ {
389
+ "epoch": 3.36,
390
+ "grad_norm": 41.267818450927734,
391
+ "learning_rate": 1.8861209964412814e-05,
392
+ "loss": 1.7655,
393
+ "step": 420
394
+ },
395
+ {
396
+ "epoch": 3.44,
397
+ "grad_norm": 41.248863220214844,
398
+ "learning_rate": 1.797153024911032e-05,
399
+ "loss": 1.8265,
400
+ "step": 430
401
+ },
402
+ {
403
+ "epoch": 3.52,
404
+ "grad_norm": 42.36342239379883,
405
+ "learning_rate": 1.708185053380783e-05,
406
+ "loss": 1.8632,
407
+ "step": 440
408
+ },
409
+ {
410
+ "epoch": 3.6,
411
+ "grad_norm": 155.6268310546875,
412
+ "learning_rate": 1.619217081850534e-05,
413
+ "loss": 1.5505,
414
+ "step": 450
415
+ },
416
+ {
417
+ "epoch": 3.68,
418
+ "grad_norm": 234.6924285888672,
419
+ "learning_rate": 1.530249110320285e-05,
420
+ "loss": 1.6754,
421
+ "step": 460
422
+ },
423
+ {
424
+ "epoch": 3.76,
425
+ "grad_norm": 42.33539581298828,
426
+ "learning_rate": 1.4412811387900358e-05,
427
+ "loss": 1.9368,
428
+ "step": 470
429
+ },
430
+ {
431
+ "epoch": 3.84,
432
+ "grad_norm": 64.73855590820312,
433
+ "learning_rate": 1.3523131672597866e-05,
434
+ "loss": 1.6528,
435
+ "step": 480
436
+ },
437
+ {
438
+ "epoch": 3.92,
439
+ "grad_norm": 26.477352142333984,
440
+ "learning_rate": 1.2633451957295376e-05,
441
+ "loss": 1.6419,
442
+ "step": 490
443
+ },
444
+ {
445
+ "epoch": 4.0,
446
+ "grad_norm": 31.107269287109375,
447
+ "learning_rate": 1.1743772241992882e-05,
448
+ "loss": 1.8633,
449
+ "step": 500
450
+ },
451
+ {
452
+ "epoch": 4.0,
453
+ "eval_loss": 1.847435474395752,
454
+ "eval_map": 0.1149,
455
+ "eval_map_50": 0.2082,
456
+ "eval_map_75": 0.1041,
457
+ "eval_map_Coverall": 0.4733,
458
+ "eval_map_Face_Shield": 0.0,
459
+ "eval_map_Gloves": 0.0509,
460
+ "eval_map_Goggles": 0.0,
461
+ "eval_map_Mask": 0.0505,
462
+ "eval_map_large": 0.1359,
463
+ "eval_map_medium": 0.0191,
464
+ "eval_map_small": 0.0683,
465
+ "eval_mar_1": 0.1124,
466
+ "eval_mar_10": 0.2271,
467
+ "eval_mar_100": 0.2731,
468
+ "eval_mar_100_Coverall": 0.7067,
469
+ "eval_mar_100_Face_Shield": 0.0,
470
+ "eval_mar_100_Gloves": 0.2492,
471
+ "eval_mar_100_Goggles": 0.0,
472
+ "eval_mar_100_Mask": 0.4096,
473
+ "eval_mar_large": 0.2379,
474
+ "eval_mar_medium": 0.1471,
475
+ "eval_mar_small": 0.1477,
476
+ "eval_runtime": 0.9243,
477
+ "eval_samples_per_second": 31.375,
478
+ "eval_steps_per_second": 2.164,
479
+ "step": 500
480
+ },
481
+ {
482
+ "epoch": 4.08,
483
+ "grad_norm": 75.24325561523438,
484
+ "learning_rate": 1.0854092526690392e-05,
485
+ "loss": 1.7156,
486
+ "step": 510
487
+ },
488
+ {
489
+ "epoch": 4.16,
490
+ "grad_norm": 19.798181533813477,
491
+ "learning_rate": 9.9644128113879e-06,
492
+ "loss": 1.8849,
493
+ "step": 520
494
+ },
495
+ {
496
+ "epoch": 4.24,
497
+ "grad_norm": 53.58675765991211,
498
+ "learning_rate": 9.07473309608541e-06,
499
+ "loss": 1.652,
500
+ "step": 530
501
+ },
502
+ {
503
+ "epoch": 4.32,
504
+ "grad_norm": 27.688901901245117,
505
+ "learning_rate": 8.185053380782918e-06,
506
+ "loss": 1.6366,
507
+ "step": 540
508
+ },
509
+ {
510
+ "epoch": 4.4,
511
+ "grad_norm": 30.249404907226562,
512
+ "learning_rate": 7.295373665480428e-06,
513
+ "loss": 1.7331,
514
+ "step": 550
515
+ },
516
+ {
517
+ "epoch": 4.48,
518
+ "grad_norm": 48.16246032714844,
519
+ "learning_rate": 6.405693950177937e-06,
520
+ "loss": 1.7585,
521
+ "step": 560
522
+ },
523
+ {
524
+ "epoch": 4.5600000000000005,
525
+ "grad_norm": 23.550582885742188,
526
+ "learning_rate": 5.516014234875446e-06,
527
+ "loss": 1.8989,
528
+ "step": 570
529
+ },
530
+ {
531
+ "epoch": 4.64,
532
+ "grad_norm": 102.61798858642578,
533
+ "learning_rate": 4.626334519572954e-06,
534
+ "loss": 1.5747,
535
+ "step": 580
536
+ },
537
+ {
538
+ "epoch": 4.72,
539
+ "grad_norm": 28.473485946655273,
540
+ "learning_rate": 3.7366548042704624e-06,
541
+ "loss": 1.7234,
542
+ "step": 590
543
+ },
544
+ {
545
+ "epoch": 4.8,
546
+ "grad_norm": 31.658267974853516,
547
+ "learning_rate": 2.8469750889679713e-06,
548
+ "loss": 1.55,
549
+ "step": 600
550
+ },
551
+ {
552
+ "epoch": 4.88,
553
+ "grad_norm": 32.104591369628906,
554
+ "learning_rate": 1.9572953736654803e-06,
555
+ "loss": 1.7035,
556
+ "step": 610
557
+ },
558
+ {
559
+ "epoch": 4.96,
560
+ "grad_norm": 35.97725296020508,
561
+ "learning_rate": 1.0676156583629894e-06,
562
+ "loss": 1.6808,
563
+ "step": 620
564
+ },
565
+ {
566
+ "epoch": 5.0,
567
+ "eval_loss": 1.7329339981079102,
568
+ "eval_map": 0.1341,
569
+ "eval_map_50": 0.2559,
570
+ "eval_map_75": 0.1179,
571
+ "eval_map_Coverall": 0.5042,
572
+ "eval_map_Face_Shield": 0.0,
573
+ "eval_map_Gloves": 0.0962,
574
+ "eval_map_Goggles": 0.0,
575
+ "eval_map_Mask": 0.07,
576
+ "eval_map_large": 0.1421,
577
+ "eval_map_medium": 0.0367,
578
+ "eval_map_small": 0.0652,
579
+ "eval_mar_1": 0.1278,
580
+ "eval_mar_10": 0.2553,
581
+ "eval_mar_100": 0.2932,
582
+ "eval_mar_100_Coverall": 0.7222,
583
+ "eval_mar_100_Face_Shield": 0.0,
584
+ "eval_mar_100_Gloves": 0.3246,
585
+ "eval_mar_100_Goggles": 0.0,
586
+ "eval_mar_100_Mask": 0.4192,
587
+ "eval_mar_large": 0.2122,
588
+ "eval_mar_medium": 0.1701,
589
+ "eval_mar_small": 0.1689,
590
+ "eval_runtime": 0.934,
591
+ "eval_samples_per_second": 31.048,
592
+ "eval_steps_per_second": 2.141,
593
+ "step": 625
594
+ }
595
+ ],
596
+ "logging_steps": 10,
597
+ "max_steps": 625,
598
+ "num_input_tokens_seen": 0,
599
+ "num_train_epochs": 5,
600
+ "save_steps": 500,
601
+ "total_flos": 1.3438569384e+18,
602
+ "train_batch_size": 8,
603
+ "trial_name": null,
604
+ "trial_params": null
605
+ }
checkpoint-625/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa74bc64d54edaec774743a0e0459391d10d4ac85c1413264ee0481492aad766
3
+ size 4984
config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/detr-resnet-50",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "relu",
5
+ "architectures": [
6
+ "DetrForObjectDetection"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "auxiliary_loss": false,
10
+ "backbone": "resnet50",
11
+ "backbone_config": null,
12
+ "backbone_kwargs": null,
13
+ "bbox_cost": 5,
14
+ "bbox_loss_coefficient": 5,
15
+ "class_cost": 1,
16
+ "classifier_dropout": 0.0,
17
+ "d_model": 256,
18
+ "decoder_attention_heads": 8,
19
+ "decoder_ffn_dim": 2048,
20
+ "decoder_layerdrop": 0.0,
21
+ "decoder_layers": 6,
22
+ "dice_loss_coefficient": 1,
23
+ "dilation": false,
24
+ "dropout": 0.1,
25
+ "encoder_attention_heads": 8,
26
+ "encoder_ffn_dim": 2048,
27
+ "encoder_layerdrop": 0.0,
28
+ "encoder_layers": 6,
29
+ "eos_coefficient": 0.1,
30
+ "giou_cost": 2,
31
+ "giou_loss_coefficient": 2,
32
+ "id2label": {
33
+ "0": "Coverall",
34
+ "1": "Face_Shield",
35
+ "2": "Gloves",
36
+ "3": "Goggles",
37
+ "4": "Mask"
38
+ },
39
+ "init_std": 0.02,
40
+ "init_xavier_std": 1.0,
41
+ "is_encoder_decoder": true,
42
+ "label2id": {
43
+ "Coverall": 0,
44
+ "Face_Shield": 1,
45
+ "Gloves": 2,
46
+ "Goggles": 3,
47
+ "Mask": 4
48
+ },
49
+ "mask_loss_coefficient": 1,
50
+ "max_position_embeddings": 1024,
51
+ "model_type": "detr",
52
+ "num_channels": 3,
53
+ "num_hidden_layers": 6,
54
+ "num_queries": 100,
55
+ "position_embedding_type": "sine",
56
+ "scale_embedding": false,
57
+ "torch_dtype": "float32",
58
+ "transformers_version": "4.40.1",
59
+ "use_pretrained_backbone": true,
60
+ "use_timm_backbone": true
61
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d39d274c25ce58eeb595dda77b61165ac29c61aa45eecf7efcb0e835f3c06d3
3
+ size 166498936
preprocessor_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_valid_processor_keys": [
3
+ "images",
4
+ "annotations",
5
+ "return_segmentation_masks",
6
+ "masks_path",
7
+ "do_resize",
8
+ "size",
9
+ "resample",
10
+ "do_rescale",
11
+ "rescale_factor",
12
+ "do_normalize",
13
+ "do_convert_annotations",
14
+ "image_mean",
15
+ "image_std",
16
+ "do_pad",
17
+ "format",
18
+ "return_tensors",
19
+ "data_format",
20
+ "input_data_format"
21
+ ],
22
+ "do_convert_annotations": true,
23
+ "do_normalize": true,
24
+ "do_pad": false,
25
+ "do_rescale": true,
26
+ "do_resize": false,
27
+ "format": "coco_detection",
28
+ "image_mean": [
29
+ 0.485,
30
+ 0.456,
31
+ 0.406
32
+ ],
33
+ "image_processor_type": "DetrImageProcessor",
34
+ "image_std": [
35
+ 0.229,
36
+ 0.224,
37
+ 0.225
38
+ ],
39
+ "resample": 2,
40
+ "rescale_factor": 0.00392156862745098,
41
+ "size": {
42
+ "longest_edge": 600
43
+ }
44
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa74bc64d54edaec774743a0e0459391d10d4ac85c1413264ee0481492aad766
3
+ size 4984
training_params.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "data_path": "cppe-5",
3
+ "model": "facebook/detr-resnet-50",
4
+ "username": "abhishek",
5
+ "lr": 5e-05,
6
+ "epochs": 5,
7
+ "batch_size": 8,
8
+ "warmup_ratio": 0.1,
9
+ "gradient_accumulation": 1,
10
+ "optimizer": "adamw_torch",
11
+ "scheduler": "linear",
12
+ "weight_decay": 0.0,
13
+ "max_grad_norm": 1.0,
14
+ "seed": 42,
15
+ "train_split": "train",
16
+ "valid_split": "test",
17
+ "logging_steps": 10,
18
+ "project_name": "autotrain-obj-det-test-1",
19
+ "auto_find_batch_size": false,
20
+ "mixed_precision": "fp16",
21
+ "save_total_limit": 1,
22
+ "push_to_hub": true,
23
+ "evaluation_strategy": "epoch",
24
+ "image_column": "image",
25
+ "objects_column": "objects",
26
+ "log": "none",
27
+ "image_square_size": 600
28
+ }