Hugging Face's logo

sam1120
/
safety-utcustom-train-SF-RGB-b5

Image Segmentation
Transformers
PyTorch
TensorBoard
segformer
vision
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
sam1120 commited on
Commit
d046ddf
•
1 Parent(s): 9dea228

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +61 -41
README.md CHANGED
@@ -1,8 +1,6 @@
1
  ---
2
  license: other
3
  tags:
4
- - vision
5
- - image-segmentation
6
  - generated_from_trainer
7
  model-index:
8
  - name: safety-utcustom-train-SF-RGB-b5
@@ -14,18 +12,18 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  # safety-utcustom-train-SF-RGB-b5
16
 
17
- This model is a fine-tuned version of [nvidia/mit-b5](https://huggingface.co/nvidia/mit-b5) on the sam1120/safety-utcustom-TRAIN dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 0.2537
20
- - Mean Iou: 0.4972
21
- - Mean Accuracy: 0.8434
22
- - Overall Accuracy: 0.9802
23
  - Accuracy Unlabeled: nan
24
- - Accuracy Safe: 0.6980
25
- - Accuracy Unsafe: 0.9888
26
  - Iou Unlabeled: 0.0
27
- - Iou Safe: 0.5118
28
- - Iou Unsafe: 0.9798
29
 
30
  ## Model description
31
 
@@ -51,7 +49,7 @@ The following hyperparameters were used during training:
51
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
  - lr_scheduler_type: linear
53
  - lr_scheduler_warmup_ratio: 0.05
54
- - num_epochs: 110
55
 
56
  ### Training results
57
 
@@ -149,35 +147,57 @@ The following hyperparameters were used during training:
149
  | 0.3287 | 81.82 | 900 | 0.6518 | nan | 0.9876 | 0.4652 | 0.0 | 0.9773 | 0.2858 | 0.8197 | 0.4808 | 0.9777 |
150
  | 0.3067 | 82.73 | 910 | 0.6528 | nan | 0.9875 | 0.4654 | 0.0 | 0.9772 | 0.2861 | 0.8202 | 0.4809 | 0.9776 |
151
  | 0.3374 | 83.64 | 920 | 0.6577 | nan | 0.9870 | 0.4631 | 0.0 | 0.9768 | 0.2869 | 0.8224 | 0.4800 | 0.9773 |
152
- | 0.3171 | 84.55 | 930 | 0.2706 | 0.4832 | 0.8164 | 0.9785 | nan | 0.6442 | 0.9887 | 0.0 | 0.4714 | 0.9781 |
153
- | 0.3156 | 85.45 | 940 | 0.2708 | 0.4849 | 0.8110 | 0.9793 | nan | 0.6321 | 0.9899 | 0.0 | 0.4757 | 0.9789 |
154
- | 0.2749 | 86.36 | 950 | 0.2760 | 0.4850 | 0.8202 | 0.9787 | nan | 0.6518 | 0.9887 | 0.0 | 0.4765 | 0.9783 |
155
- | 0.2725 | 87.27 | 960 | 0.2780 | 0.4847 | 0.8279 | 0.9782 | nan | 0.6681 | 0.9876 | 0.0 | 0.4765 | 0.9777 |
156
- | 0.2948 | 88.18 | 970 | 0.2636 | 0.4878 | 0.8228 | 0.9793 | nan | 0.6565 | 0.9891 | 0.0 | 0.4845 | 0.9788 |
157
- | 0.2972 | 89.09 | 980 | 0.2770 | 0.4870 | 0.8301 | 0.9786 | nan | 0.6722 | 0.9879 | 0.0 | 0.4829 | 0.9782 |
158
- | 0.3101 | 90.0 | 990 | 0.2765 | 0.4881 | 0.8297 | 0.9788 | nan | 0.6711 | 0.9882 | 0.0 | 0.4859 | 0.9784 |
159
- | 0.2874 | 90.91 | 1000 | 0.2690 | 0.4896 | 0.8288 | 0.9793 | nan | 0.6689 | 0.9888 | 0.0 | 0.4899 | 0.9789 |
160
- | 0.275 | 91.82 | 1010 | 0.2593 | 0.4915 | 0.8221 | 0.9802 | nan | 0.6542 | 0.9901 | 0.0 | 0.4947 | 0.9798 |
161
- | 0.2711 | 92.73 | 1020 | 0.2608 | 0.4917 | 0.8283 | 0.9798 | nan | 0.6673 | 0.9893 | 0.0 | 0.4957 | 0.9794 |
162
- | 0.2691 | 93.64 | 1030 | 0.2609 | 0.4915 | 0.8352 | 0.9794 | nan | 0.6819 | 0.9884 | 0.0 | 0.4954 | 0.9789 |
163
- | 0.274 | 94.55 | 1040 | 0.2542 | 0.4935 | 0.8309 | 0.9801 | nan | 0.6722 | 0.9895 | 0.0 | 0.5007 | 0.9797 |
164
- | 0.27 | 95.45 | 1050 | 0.2357 | 0.4966 | 0.8177 | 0.9816 | nan | 0.6436 | 0.9919 | 0.0 | 0.5087 | 0.9812 |
165
- | 0.255 | 96.36 | 1060 | 0.2460 | 0.4959 | 0.8287 | 0.9808 | nan | 0.6671 | 0.9903 | 0.0 | 0.5074 | 0.9804 |
166
- | 0.2756 | 97.27 | 1070 | 0.2411 | 0.4974 | 0.8271 | 0.9812 | nan | 0.6634 | 0.9909 | 0.0 | 0.5113 | 0.9809 |
167
- | 0.2473 | 98.18 | 1080 | 0.2447 | 0.4974 | 0.8314 | 0.9810 | nan | 0.6724 | 0.9904 | 0.0 | 0.5116 | 0.9806 |
168
- | 0.266 | 99.09 | 1090 | 0.2455 | 0.4974 | 0.8339 | 0.9808 | nan | 0.6778 | 0.9901 | 0.0 | 0.5118 | 0.9804 |
169
- | 0.2682 | 100.0 | 1100 | 0.2359 | 0.4989 | 0.8281 | 0.9815 | nan | 0.6651 | 0.9911 | 0.0 | 0.5156 | 0.9811 |
170
- | 0.2607 | 100.91 | 1110 | 0.2376 | 0.4984 | 0.8322 | 0.9812 | nan | 0.6739 | 0.9905 | 0.0 | 0.5144 | 0.9808 |
171
- | 0.2506 | 101.82 | 1120 | 0.2380 | 0.4987 | 0.8317 | 0.9813 | nan | 0.6727 | 0.9907 | 0.0 | 0.5153 | 0.9809 |
172
- | 0.2729 | 102.73 | 1130 | 0.2375 | 0.4984 | 0.8352 | 0.9810 | nan | 0.6802 | 0.9902 | 0.0 | 0.5146 | 0.9806 |
173
- | 0.2348 | 103.64 | 1140 | 0.2335 | 0.4994 | 0.8319 | 0.9814 | nan | 0.6731 | 0.9908 | 0.0 | 0.5172 | 0.9810 |
174
- | 0.2409 | 104.55 | 1150 | 0.2385 | 0.4991 | 0.8343 | 0.9812 | nan | 0.6781 | 0.9904 | 0.0 | 0.5164 | 0.9808 |
175
- | 0.2737 | 105.45 | 1160 | 0.2357 | 0.4995 | 0.8340 | 0.9813 | nan | 0.6774 | 0.9906 | 0.0 | 0.5177 | 0.9809 |
176
- | 0.2857 | 106.36 | 1170 | 0.2304 | 0.5006 | 0.8287 | 0.9818 | nan | 0.6659 | 0.9915 | 0.0 | 0.5203 | 0.9815 |
177
- | 0.2503 | 107.27 | 1180 | 0.2397 | 0.4993 | 0.8362 | 0.9811 | nan | 0.6822 | 0.9902 | 0.0 | 0.5172 | 0.9807 |
178
- | 0.2524 | 108.18 | 1190 | 0.2294 | 0.5004 | 0.8325 | 0.9816 | nan | 0.6741 | 0.9909 | 0.0 | 0.5199 | 0.9812 |
179
- | 0.2722 | 109.09 | 1200 | 0.2461 | 0.4982 | 0.8411 | 0.9806 | nan | 0.6928 | 0.9894 | 0.0 | 0.5145 | 0.9802 |
180
- | 0.2777 | 110.0 | 1210 | 0.2537 | 0.4972 | 0.8434 | 0.9802 | nan | 0.6980 | 0.9888 | 0.0 | 0.5118 | 0.9798 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
181
 
182
 
183
  ### Framework versions
 
1
  ---
2
  license: other
3
  tags:
 
 
4
  - generated_from_trainer
5
  model-index:
6
  - name: safety-utcustom-train-SF-RGB-b5
 
12
 
13
  # safety-utcustom-train-SF-RGB-b5
14
 
15
+ This model is a fine-tuned version of [nvidia/mit-b5](https://huggingface.co/nvidia/mit-b5) on the None dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Loss: 0.2236
18
+ - Mean Iou: 0.5051
19
+ - Mean Accuracy: 0.8441
20
+ - Overall Accuracy: 0.9820
21
  - Accuracy Unlabeled: nan
22
+ - Accuracy Safe: 0.6977
23
+ - Accuracy Unsafe: 0.9906
24
  - Iou Unlabeled: 0.0
25
+ - Iou Safe: 0.5337
26
+ - Iou Unsafe: 0.9816
27
 
28
  ## Model description
29
 
 
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: linear
51
  - lr_scheduler_warmup_ratio: 0.05
52
+ - num_epochs: 130
53
 
54
  ### Training results
55
 
 
147
  | 0.3287 | 81.82 | 900 | 0.6518 | nan | 0.9876 | 0.4652 | 0.0 | 0.9773 | 0.2858 | 0.8197 | 0.4808 | 0.9777 |
148
  | 0.3067 | 82.73 | 910 | 0.6528 | nan | 0.9875 | 0.4654 | 0.0 | 0.9772 | 0.2861 | 0.8202 | 0.4809 | 0.9776 |
149
  | 0.3374 | 83.64 | 920 | 0.6577 | nan | 0.9870 | 0.4631 | 0.0 | 0.9768 | 0.2869 | 0.8224 | 0.4800 | 0.9773 |
150
+ | 0.3171 | 84.55 | 930 | 0.6442 | nan | 0.9887 | 0.4714 | 0.0 | 0.9781 | 0.2706 | 0.8164 | 0.4832 | 0.9785 |
151
+ | 0.3156 | 85.45 | 940 | 0.6321 | nan | 0.9899 | 0.4757 | 0.0 | 0.9789 | 0.2708 | 0.8110 | 0.4849 | 0.9793 |
152
+ | 0.2749 | 86.36 | 950 | 0.6518 | nan | 0.9887 | 0.4765 | 0.0 | 0.9783 | 0.2760 | 0.8202 | 0.4850 | 0.9787 |
153
+ | 0.2725 | 87.27 | 960 | 0.6681 | nan | 0.9876 | 0.4765 | 0.0 | 0.9777 | 0.2780 | 0.8279 | 0.4847 | 0.9782 |
154
+ | 0.2948 | 88.18 | 970 | 0.6565 | nan | 0.9891 | 0.4845 | 0.0 | 0.9788 | 0.2636 | 0.8228 | 0.4878 | 0.9793 |
155
+ | 0.2972 | 89.09 | 980 | 0.6722 | nan | 0.9879 | 0.4829 | 0.0 | 0.9782 | 0.2770 | 0.8301 | 0.4870 | 0.9786 |
156
+ | 0.3101 | 90.0 | 990 | 0.6711 | nan | 0.9882 | 0.4859 | 0.0 | 0.9784 | 0.2765 | 0.8297 | 0.4881 | 0.9788 |
157
+ | 0.2874 | 90.91 | 1000 | 0.6689 | nan | 0.9888 | 0.4899 | 0.0 | 0.9789 | 0.2690 | 0.8288 | 0.4896 | 0.9793 |
158
+ | 0.275 | 91.82 | 1010 | 0.6542 | nan | 0.9901 | 0.4947 | 0.0 | 0.9798 | 0.2593 | 0.8221 | 0.4915 | 0.9802 |
159
+ | 0.2711 | 92.73 | 1020 | 0.6673 | nan | 0.9893 | 0.4957 | 0.0 | 0.9794 | 0.2608 | 0.8283 | 0.4917 | 0.9798 |
160
+ | 0.2691 | 93.64 | 1030 | 0.6819 | nan | 0.9884 | 0.4954 | 0.0 | 0.9789 | 0.2609 | 0.8352 | 0.4915 | 0.9794 |
161
+ | 0.274 | 94.55 | 1040 | 0.6722 | nan | 0.9895 | 0.5007 | 0.0 | 0.9797 | 0.2542 | 0.8309 | 0.4935 | 0.9801 |
162
+ | 0.27 | 95.45 | 1050 | 0.6436 | nan | 0.9919 | 0.5087 | 0.0 | 0.9812 | 0.2357 | 0.8177 | 0.4966 | 0.9816 |
163
+ | 0.255 | 96.36 | 1060 | 0.6671 | nan | 0.9903 | 0.5074 | 0.0 | 0.9804 | 0.2460 | 0.8287 | 0.4959 | 0.9808 |
164
+ | 0.2756 | 97.27 | 1070 | 0.6634 | nan | 0.9909 | 0.5113 | 0.0 | 0.9809 | 0.2411 | 0.8271 | 0.4974 | 0.9812 |
165
+ | 0.2473 | 98.18 | 1080 | 0.6724 | nan | 0.9904 | 0.5116 | 0.0 | 0.9806 | 0.2447 | 0.8314 | 0.4974 | 0.9810 |
166
+ | 0.266 | 99.09 | 1090 | 0.6778 | nan | 0.9901 | 0.5118 | 0.0 | 0.9804 | 0.2455 | 0.8339 | 0.4974 | 0.9808 |
167
+ | 0.2682 | 100.0 | 1100 | 0.6651 | nan | 0.9911 | 0.5156 | 0.0 | 0.9811 | 0.2359 | 0.8281 | 0.4989 | 0.9815 |
168
+ | 0.2607 | 100.91 | 1110 | 0.6739 | nan | 0.9905 | 0.5144 | 0.0 | 0.9808 | 0.2376 | 0.8322 | 0.4984 | 0.9812 |
169
+ | 0.2506 | 101.82 | 1120 | 0.6727 | nan | 0.9907 | 0.5153 | 0.0 | 0.9809 | 0.2380 | 0.8317 | 0.4987 | 0.9813 |
170
+ | 0.2729 | 102.73 | 1130 | 0.6802 | nan | 0.9902 | 0.5146 | 0.0 | 0.9806 | 0.2375 | 0.8352 | 0.4984 | 0.9810 |
171
+ | 0.2348 | 103.64 | 1140 | 0.6731 | nan | 0.9908 | 0.5172 | 0.0 | 0.9810 | 0.2335 | 0.8319 | 0.4994 | 0.9814 |
172
+ | 0.2409 | 104.55 | 1150 | 0.6781 | nan | 0.9904 | 0.5164 | 0.0 | 0.9808 | 0.2385 | 0.8343 | 0.4991 | 0.9812 |
173
+ | 0.2737 | 105.45 | 1160 | 0.6774 | nan | 0.9906 | 0.5177 | 0.0 | 0.9809 | 0.2357 | 0.8340 | 0.4995 | 0.9813 |
174
+ | 0.2857 | 106.36 | 1170 | 0.6659 | nan | 0.9915 | 0.5203 | 0.0 | 0.9815 | 0.2304 | 0.8287 | 0.5006 | 0.9818 |
175
+ | 0.2503 | 107.27 | 1180 | 0.6822 | nan | 0.9902 | 0.5172 | 0.0 | 0.9807 | 0.2397 | 0.8362 | 0.4993 | 0.9811 |
176
+ | 0.2524 | 108.18 | 1190 | 0.6741 | nan | 0.9909 | 0.5199 | 0.0 | 0.9812 | 0.2294 | 0.8325 | 0.5004 | 0.9816 |
177
+ | 0.2722 | 109.09 | 1200 | 0.6928 | nan | 0.9894 | 0.5145 | 0.0 | 0.9802 | 0.2461 | 0.8411 | 0.4982 | 0.9806 |
178
+ | 0.2761 | 110.0 | 1210 | 0.2530 | 0.4979 | 0.8435 | 0.9804 | nan | 0.6981 | 0.9889 | 0.0 | 0.5136 | 0.9799 |
179
+ | 0.2595 | 110.91 | 1220 | 0.2386 | 0.5002 | 0.8395 | 0.9811 | nan | 0.6891 | 0.9900 | 0.0 | 0.5197 | 0.9808 |
180
+ | 0.2877 | 111.82 | 1230 | 0.2260 | 0.5022 | 0.8345 | 0.9819 | nan | 0.6779 | 0.9911 | 0.0 | 0.5251 | 0.9815 |
181
+ | 0.2307 | 112.73 | 1240 | 0.2219 | 0.5031 | 0.8317 | 0.9822 | nan | 0.6717 | 0.9916 | 0.0 | 0.5275 | 0.9818 |
182
+ | 0.2559 | 113.64 | 1250 | 0.2382 | 0.5011 | 0.8413 | 0.9812 | nan | 0.6927 | 0.9900 | 0.0 | 0.5224 | 0.9808 |
183
+ | 0.2668 | 114.55 | 1260 | 0.2275 | 0.5024 | 0.8395 | 0.9816 | nan | 0.6883 | 0.9906 | 0.0 | 0.5258 | 0.9813 |
184
+ | 0.2269 | 115.45 | 1270 | 0.2212 | 0.5038 | 0.8356 | 0.9821 | nan | 0.6798 | 0.9913 | 0.0 | 0.5296 | 0.9818 |
185
+ | 0.2713 | 116.36 | 1280 | 0.2265 | 0.5041 | 0.8365 | 0.9821 | nan | 0.6817 | 0.9913 | 0.0 | 0.5305 | 0.9818 |
186
+ | 0.2351 | 117.27 | 1290 | 0.2113 | 0.5064 | 0.8255 | 0.9832 | nan | 0.6580 | 0.9931 | 0.0 | 0.5363 | 0.9828 |
187
+ | 0.2306 | 118.18 | 1300 | 0.2258 | 0.5053 | 0.8354 | 0.9824 | nan | 0.6791 | 0.9917 | 0.0 | 0.5337 | 0.9821 |
188
+ | 0.2371 | 119.09 | 1310 | 0.2234 | 0.5057 | 0.8342 | 0.9826 | nan | 0.6765 | 0.9919 | 0.0 | 0.5348 | 0.9822 |
189
+ | 0.277 | 120.0 | 1320 | 0.2263 | 0.5061 | 0.8340 | 0.9827 | nan | 0.6760 | 0.9920 | 0.0 | 0.5359 | 0.9823 |
190
+ | 0.2121 | 120.91 | 1330 | 0.2191 | 0.5065 | 0.8337 | 0.9828 | nan | 0.6752 | 0.9921 | 0.0 | 0.5370 | 0.9824 |
191
+ | 0.2426 | 121.82 | 1340 | 0.2237 | 0.5055 | 0.8400 | 0.9822 | nan | 0.6889 | 0.9912 | 0.0 | 0.5345 | 0.9819 |
192
+ | 0.2429 | 122.73 | 1350 | 0.2106 | 0.5069 | 0.8340 | 0.9829 | nan | 0.6757 | 0.9922 | 0.0 | 0.5383 | 0.9825 |
193
+ | 0.244 | 123.64 | 1360 | 0.2133 | 0.5062 | 0.8395 | 0.9824 | nan | 0.6876 | 0.9914 | 0.0 | 0.5364 | 0.9821 |
194
+ | 0.2287 | 124.55 | 1370 | 0.2126 | 0.5068 | 0.8371 | 0.9827 | nan | 0.6823 | 0.9918 | 0.0 | 0.5381 | 0.9823 |
195
+ | 0.2223 | 125.45 | 1380 | 0.1988 | 0.5083 | 0.8305 | 0.9833 | nan | 0.6681 | 0.9929 | 0.0 | 0.5420 | 0.9830 |
196
+ | 0.2275 | 126.36 | 1390 | 0.2164 | 0.5061 | 0.8409 | 0.9823 | nan | 0.6906 | 0.9912 | 0.0 | 0.5363 | 0.9820 |
197
+ | 0.2378 | 127.27 | 1400 | 0.2201 | 0.5062 | 0.8405 | 0.9824 | nan | 0.6898 | 0.9913 | 0.0 | 0.5365 | 0.9820 |
198
+ | 0.2385 | 128.18 | 1410 | 0.2286 | 0.5046 | 0.8455 | 0.9818 | nan | 0.7006 | 0.9903 | 0.0 | 0.5324 | 0.9814 |
199
+ | 0.2787 | 129.09 | 1420 | 0.2157 | 0.5063 | 0.8405 | 0.9824 | nan | 0.6896 | 0.9913 | 0.0 | 0.5369 | 0.9820 |
200
+ | 0.2377 | 130.0 | 1430 | 0.2236 | 0.5051 | 0.8441 | 0.9820 | nan | 0.6977 | 0.9906 | 0.0 | 0.5337 | 0.9816 |
201
 
202
 
203
  ### Framework versions