Hugging Face's logo

AmirGenAI
/
segformer-b0-finetuned-wrinkle

Image Segmentation
Transformers
TensorBoard
Safetensors
segformer
vision
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
AmirGenAI commited on
Commit
e801424
·
verified ·
1 Parent(s): 4edaa6a

End of training

Browse files
Files changed (4) hide show
  1. README.md +14 -106
  2. model.safetensors +1 -1
  3. runs/Dec03_08-42-52_a915ad9a3662/events.out.tfevents.1733215381.a915ad9a3662.211.0 +3 -0
  4. training_args.bin +1 -1
README.md CHANGED
@@ -18,14 +18,19 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on the face-wrinkles dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.0213
22
- - Mean Iou: 0.0905
23
- - Mean Accuracy: 0.1810
24
- - Overall Accuracy: 0.1810
25
- - Accuracy Unlabeled: nan
26
- - Accuracy Wrinkle: 0.1810
27
- - Iou Unlabeled: 0.0
28
- - Iou Wrinkle: 0.1810
 
 
 
 
 
29
 
30
  ## Model description
31
 
@@ -50,104 +55,7 @@ The following hyperparameters were used during training:
50
  - seed: 42
51
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
52
  - lr_scheduler_type: linear
53
- - num_epochs: 5
54
-
55
- ### Training results
56
-
57
- | Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy Unlabeled | Accuracy Wrinkle | Iou Unlabeled | Iou Wrinkle |
58
- |:-------------:|:------:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:------------------:|:----------------:|:-------------:|:-----------:|
59
- | 0.1648 | 0.0545 | 20 | 0.1431 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
60
- | 0.1387 | 0.1090 | 40 | 0.1201 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
61
- | 0.1233 | 0.1635 | 60 | 0.0945 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
62
- | 0.0931 | 0.2180 | 80 | 0.0993 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
63
- | 0.0883 | 0.2725 | 100 | 0.0759 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
64
- | 0.0686 | 0.3270 | 120 | 0.0663 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
65
- | 0.067 | 0.3815 | 140 | 0.0621 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
66
- | 0.0564 | 0.4360 | 160 | 0.0557 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
67
- | 0.0577 | 0.4905 | 180 | 0.0487 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
68
- | 0.0436 | 0.5450 | 200 | 0.0466 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
69
- | 0.0585 | 0.5995 | 220 | 0.0455 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
70
- | 0.0589 | 0.6540 | 240 | 0.0427 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
71
- | 0.0391 | 0.7084 | 260 | 0.0403 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
72
- | 0.0405 | 0.7629 | 280 | 0.0404 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
73
- | 0.0304 | 0.8174 | 300 | 0.0390 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
74
- | 0.0306 | 0.8719 | 320 | 0.0362 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
75
- | 0.0251 | 0.9264 | 340 | 0.0341 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
76
- | 0.0226 | 0.9809 | 360 | 0.0332 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
77
- | 0.0447 | 1.0354 | 380 | 0.0320 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
78
- | 0.0338 | 1.0899 | 400 | 0.0317 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
79
- | 0.0272 | 1.1444 | 420 | 0.0303 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
80
- | 0.0296 | 1.1989 | 440 | 0.0342 | 0.0054 | 0.0109 | 0.0109 | nan | 0.0109 | 0.0 | 0.0109 |
81
- | 0.0361 | 1.2534 | 460 | 0.0298 | 0.0 | 0.0 | 0.0 | nan | 0.0 | 0.0 | 0.0 |
82
- | 0.0277 | 1.3079 | 480 | 0.0296 | 0.0045 | 0.0090 | 0.0090 | nan | 0.0090 | 0.0 | 0.0090 |
83
- | 0.0247 | 1.3624 | 500 | 0.0283 | 0.0005 | 0.0009 | 0.0009 | nan | 0.0009 | 0.0 | 0.0009 |
84
- | 0.0214 | 1.4169 | 520 | 0.0284 | 0.0016 | 0.0031 | 0.0031 | nan | 0.0031 | 0.0 | 0.0031 |
85
- | 0.0301 | 1.4714 | 540 | 0.0275 | 0.0139 | 0.0277 | 0.0277 | nan | 0.0277 | 0.0 | 0.0277 |
86
- | 0.0245 | 1.5259 | 560 | 0.0266 | 0.0059 | 0.0117 | 0.0117 | nan | 0.0117 | 0.0 | 0.0117 |
87
- | 0.0206 | 1.5804 | 580 | 0.0291 | 0.0399 | 0.0798 | 0.0798 | nan | 0.0798 | 0.0 | 0.0798 |
88
- | 0.0426 | 1.6349 | 600 | 0.0269 | 0.0035 | 0.0070 | 0.0070 | nan | 0.0070 | 0.0 | 0.0070 |
89
- | 0.0351 | 1.6894 | 620 | 0.0272 | 0.0157 | 0.0314 | 0.0314 | nan | 0.0314 | 0.0 | 0.0314 |
90
- | 0.0503 | 1.7439 | 640 | 0.0266 | 0.0085 | 0.0170 | 0.0170 | nan | 0.0170 | 0.0 | 0.0170 |
91
- | 0.0381 | 1.7984 | 660 | 0.0261 | 0.0016 | 0.0033 | 0.0033 | nan | 0.0033 | 0.0 | 0.0033 |
92
- | 0.0307 | 1.8529 | 680 | 0.0261 | 0.0068 | 0.0136 | 0.0136 | nan | 0.0136 | 0.0 | 0.0136 |
93
- | 0.0266 | 1.9074 | 700 | 0.0248 | 0.0087 | 0.0174 | 0.0174 | nan | 0.0174 | 0.0 | 0.0174 |
94
- | 0.0336 | 1.9619 | 720 | 0.0250 | 0.0054 | 0.0109 | 0.0109 | nan | 0.0109 | 0.0 | 0.0109 |
95
- | 0.0263 | 2.0163 | 740 | 0.0246 | 0.0040 | 0.0080 | 0.0080 | nan | 0.0080 | 0.0 | 0.0080 |
96
- | 0.0249 | 2.0708 | 760 | 0.0245 | 0.0133 | 0.0267 | 0.0267 | nan | 0.0267 | 0.0 | 0.0267 |
97
- | 0.0196 | 2.1253 | 780 | 0.0240 | 0.0212 | 0.0424 | 0.0424 | nan | 0.0424 | 0.0 | 0.0424 |
98
- | 0.0289 | 2.1798 | 800 | 0.0247 | 0.0402 | 0.0804 | 0.0804 | nan | 0.0804 | 0.0 | 0.0804 |
99
- | 0.0211 | 2.2343 | 820 | 0.0240 | 0.0218 | 0.0436 | 0.0436 | nan | 0.0436 | 0.0 | 0.0436 |
100
- | 0.0183 | 2.2888 | 840 | 0.0244 | 0.0324 | 0.0647 | 0.0647 | nan | 0.0647 | 0.0 | 0.0647 |
101
- | 0.0394 | 2.3433 | 860 | 0.0245 | 0.0495 | 0.0991 | 0.0991 | nan | 0.0991 | 0.0 | 0.0991 |
102
- | 0.0242 | 2.3978 | 880 | 0.0239 | 0.0096 | 0.0193 | 0.0193 | nan | 0.0193 | 0.0 | 0.0193 |
103
- | 0.0349 | 2.4523 | 900 | 0.0231 | 0.0207 | 0.0414 | 0.0414 | nan | 0.0414 | 0.0 | 0.0414 |
104
- | 0.0141 | 2.5068 | 920 | 0.0229 | 0.0264 | 0.0528 | 0.0528 | nan | 0.0528 | 0.0 | 0.0528 |
105
- | 0.0134 | 2.5613 | 940 | 0.0234 | 0.0559 | 0.1117 | 0.1117 | nan | 0.1117 | 0.0 | 0.1117 |
106
- | 0.0159 | 2.6158 | 960 | 0.0231 | 0.0227 | 0.0453 | 0.0453 | nan | 0.0453 | 0.0 | 0.0453 |
107
- | 0.0179 | 2.6703 | 980 | 0.0232 | 0.0139 | 0.0278 | 0.0278 | nan | 0.0278 | 0.0 | 0.0278 |
108
- | 0.0159 | 2.7248 | 1000 | 0.0235 | 0.0339 | 0.0677 | 0.0677 | nan | 0.0677 | 0.0 | 0.0677 |
109
- | 0.0115 | 2.7793 | 1020 | 0.0226 | 0.0246 | 0.0492 | 0.0492 | nan | 0.0492 | 0.0 | 0.0492 |
110
- | 0.0143 | 2.8338 | 1040 | 0.0230 | 0.0416 | 0.0832 | 0.0832 | nan | 0.0832 | 0.0 | 0.0832 |
111
- | 0.0152 | 2.8883 | 1060 | 0.0229 | 0.0719 | 0.1439 | 0.1439 | nan | 0.1439 | 0.0 | 0.1439 |
112
- | 0.0228 | 2.9428 | 1080 | 0.0228 | 0.0364 | 0.0728 | 0.0728 | nan | 0.0728 | 0.0 | 0.0728 |
113
- | 0.0283 | 2.9973 | 1100 | 0.0242 | 0.1143 | 0.2286 | 0.2286 | nan | 0.2286 | 0.0 | 0.2286 |
114
- | 0.0263 | 3.0518 | 1120 | 0.0226 | 0.0626 | 0.1252 | 0.1252 | nan | 0.1252 | 0.0 | 0.1252 |
115
- | 0.0333 | 3.1063 | 1140 | 0.0225 | 0.0556 | 0.1111 | 0.1111 | nan | 0.1111 | 0.0 | 0.1111 |
116
- | 0.0286 | 3.1608 | 1160 | 0.0223 | 0.0646 | 0.1293 | 0.1293 | nan | 0.1293 | 0.0 | 0.1293 |
117
- | 0.0177 | 3.2153 | 1180 | 0.0227 | 0.1126 | 0.2252 | 0.2252 | nan | 0.2252 | 0.0 | 0.2252 |
118
- | 0.0362 | 3.2698 | 1200 | 0.0229 | 0.0348 | 0.0695 | 0.0695 | nan | 0.0695 | 0.0 | 0.0695 |
119
- | 0.0257 | 3.3243 | 1220 | 0.0223 | 0.0835 | 0.1670 | 0.1670 | nan | 0.1670 | 0.0 | 0.1670 |
120
- | 0.0386 | 3.3787 | 1240 | 0.0223 | 0.0718 | 0.1437 | 0.1437 | nan | 0.1437 | 0.0 | 0.1437 |
121
- | 0.0282 | 3.4332 | 1260 | 0.0219 | 0.0619 | 0.1237 | 0.1237 | nan | 0.1237 | 0.0 | 0.1237 |
122
- | 0.0088 | 3.4877 | 1280 | 0.0218 | 0.0831 | 0.1662 | 0.1662 | nan | 0.1662 | 0.0 | 0.1662 |
123
- | 0.0166 | 3.5422 | 1300 | 0.0222 | 0.0814 | 0.1629 | 0.1629 | nan | 0.1629 | 0.0 | 0.1629 |
124
- | 0.0174 | 3.5967 | 1320 | 0.0222 | 0.0469 | 0.0938 | 0.0938 | nan | 0.0938 | 0.0 | 0.0938 |
125
- | 0.0191 | 3.6512 | 1340 | 0.0226 | 0.0954 | 0.1908 | 0.1908 | nan | 0.1908 | 0.0 | 0.1908 |
126
- | 0.0219 | 3.7057 | 1360 | 0.0219 | 0.1125 | 0.2250 | 0.2250 | nan | 0.2250 | 0.0 | 0.2250 |
127
- | 0.0131 | 3.7602 | 1380 | 0.0218 | 0.0866 | 0.1732 | 0.1732 | nan | 0.1732 | 0.0 | 0.1732 |
128
- | 0.0236 | 3.8147 | 1400 | 0.0217 | 0.0791 | 0.1583 | 0.1583 | nan | 0.1583 | 0.0 | 0.1583 |
129
- | 0.0268 | 3.8692 | 1420 | 0.0222 | 0.1254 | 0.2507 | 0.2507 | nan | 0.2507 | 0.0 | 0.2507 |
130
- | 0.0262 | 3.9237 | 1440 | 0.0214 | 0.0884 | 0.1768 | 0.1768 | nan | 0.1768 | 0.0 | 0.1768 |
131
- | 0.0088 | 3.9782 | 1460 | 0.0216 | 0.0956 | 0.1912 | 0.1912 | nan | 0.1912 | 0.0 | 0.1912 |
132
- | 0.0168 | 4.0327 | 1480 | 0.0216 | 0.1129 | 0.2257 | 0.2257 | nan | 0.2257 | 0.0 | 0.2257 |
133
- | 0.0085 | 4.0872 | 1500 | 0.0216 | 0.1111 | 0.2223 | 0.2223 | nan | 0.2223 | 0.0 | 0.2223 |
134
- | 0.0064 | 4.1417 | 1520 | 0.0214 | 0.0703 | 0.1405 | 0.1405 | nan | 0.1405 | 0.0 | 0.1405 |
135
- | 0.029 | 4.1962 | 1540 | 0.0217 | 0.0648 | 0.1297 | 0.1297 | nan | 0.1297 | 0.0 | 0.1297 |
136
- | 0.0114 | 4.2507 | 1560 | 0.0217 | 0.1078 | 0.2155 | 0.2155 | nan | 0.2155 | 0.0 | 0.2155 |
137
- | 0.0184 | 4.3052 | 1580 | 0.0212 | 0.1021 | 0.2042 | 0.2042 | nan | 0.2042 | 0.0 | 0.2042 |
138
- | 0.0133 | 4.3597 | 1600 | 0.0214 | 0.0877 | 0.1755 | 0.1755 | nan | 0.1755 | 0.0 | 0.1755 |
139
- | 0.0273 | 4.4142 | 1620 | 0.0215 | 0.1150 | 0.2299 | 0.2299 | nan | 0.2299 | 0.0 | 0.2299 |
140
- | 0.0152 | 4.4687 | 1640 | 0.0211 | 0.1114 | 0.2228 | 0.2228 | nan | 0.2228 | 0.0 | 0.2228 |
141
- | 0.013 | 4.5232 | 1660 | 0.0211 | 0.0851 | 0.1702 | 0.1702 | nan | 0.1702 | 0.0 | 0.1702 |
142
- | 0.0135 | 4.5777 | 1680 | 0.0211 | 0.1022 | 0.2044 | 0.2044 | nan | 0.2044 | 0.0 | 0.2044 |
143
- | 0.0307 | 4.6322 | 1700 | 0.0211 | 0.1015 | 0.2030 | 0.2030 | nan | 0.2030 | 0.0 | 0.2030 |
144
- | 0.0238 | 4.6866 | 1720 | 0.0211 | 0.1131 | 0.2262 | 0.2262 | nan | 0.2262 | 0.0 | 0.2262 |
145
- | 0.0222 | 4.7411 | 1740 | 0.0213 | 0.1335 | 0.2669 | 0.2669 | nan | 0.2669 | 0.0 | 0.2669 |
146
- | 0.0142 | 4.7956 | 1760 | 0.0212 | 0.1062 | 0.2125 | 0.2125 | nan | 0.2125 | 0.0 | 0.2125 |
147
- | 0.0196 | 4.8501 | 1780 | 0.0212 | 0.0905 | 0.1810 | 0.1810 | nan | 0.1810 | 0.0 | 0.1810 |
148
- | 0.0158 | 4.9046 | 1800 | 0.0213 | 0.1024 | 0.2048 | 0.2048 | nan | 0.2048 | 0.0 | 0.2048 |
149
- | 0.0312 | 4.9591 | 1820 | 0.0213 | 0.0905 | 0.1810 | 0.1810 | nan | 0.1810 | 0.0 | 0.1810 |
150
-
151
 
152
  ### Framework versions
153
 
 
18
 
19
  This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on the face-wrinkles dataset.
20
  It achieves the following results on the evaluation set:
21
+ - eval_loss: 0.0185
22
+ - eval_mean_iou: 0.1947
23
+ - eval_mean_accuracy: 0.3894
24
+ - eval_overall_accuracy: 0.3894
25
+ - eval_accuracy_unlabeled: nan
26
+ - eval_accuracy_wrinkle: 0.3894
27
+ - eval_iou_unlabeled: 0.0
28
+ - eval_iou_wrinkle: 0.3894
29
+ - eval_runtime: 13.3525
30
+ - eval_samples_per_second: 9.811
31
+ - eval_steps_per_second: 4.943
32
+ - epoch: 8.9918
33
+ - step: 3300
34
 
35
  ## Model description
36
 
 
55
  - seed: 42
56
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
57
  - lr_scheduler_type: linear
58
+ - num_epochs: 20
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
59
 
60
  ### Framework versions
61
 
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0b9161f2bc89d5669bc2adfd1a3436172c5b3964978682ebea62ddf7472cd9fa
3
  size 14884776
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f62fd4948e62e25b63f85f4f723ae07eedbfe5669abd4375eac9d44040c74f53
3
  size 14884776
runs/Dec03_08-42-52_a915ad9a3662/events.out.tfevents.1733215381.a915ad9a3662.211.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd5e2f04f5bb21ce47fda4ce5e63d520a6393c2844f7ae25767737343399b2e3
3
+ size 816749
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:904a4f13eb58716805ebce90795a7cd27f19291a77b19e7b2d80e06b65d7405b
3
  size 5368
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b0b4e6f49088e2730af09d71a3febd260c16b85d05603f855aca67c245afa7ec
3
  size 5368