HorcruxNo13
commited on
Commit
•
693ebf3
1
Parent(s):
5f543c3
update model card README.md
Browse files
README.md
CHANGED
@@ -1,6 +1,8 @@
|
|
1 |
---
|
2 |
license: other
|
3 |
tags:
|
|
|
|
|
4 |
- generated_from_trainer
|
5 |
model-index:
|
6 |
- name: segformer-b0-finetuned-segments-toolwear
|
@@ -12,18 +14,18 @@ should probably proofread and complete it, then remove this comment. -->
|
|
12 |
|
13 |
# segformer-b0-finetuned-segments-toolwear
|
14 |
|
15 |
-
This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on
|
16 |
It achieves the following results on the evaluation set:
|
17 |
-
- Loss: 0.
|
18 |
-
- Mean Iou: 0.
|
19 |
-
- Mean Accuracy: 0.
|
20 |
-
- Overall Accuracy: 0.
|
21 |
- Accuracy Unlabeled: nan
|
22 |
- Accuracy Tool: nan
|
23 |
-
- Accuracy Wear: 0.
|
24 |
- Iou Unlabeled: 0.0
|
25 |
- Iou Tool: nan
|
26 |
-
- Iou Wear: 0.
|
27 |
|
28 |
## Model description
|
29 |
|
@@ -54,53 +56,53 @@ The following hyperparameters were used during training:
|
|
54 |
|
55 |
| Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy Unlabeled | Accuracy Tool | Accuracy Wear | Iou Unlabeled | Iou Tool | Iou Wear |
|
56 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:------------------:|:-------------:|:-------------:|:-------------:|:--------:|:--------:|
|
57 |
-
| 0.
|
58 |
-
| 0.
|
59 |
-
| 0.
|
60 |
-
| 0.
|
61 |
-
| 0.
|
62 |
-
| 0.
|
63 |
-
| 0.
|
64 |
-
| 0.
|
65 |
-
| 0.
|
66 |
-
| 0.
|
67 |
-
| 0.
|
68 |
-
| 0.
|
69 |
-
| 0.
|
70 |
-
| 0.
|
71 |
-
| 0.
|
72 |
-
| 0.
|
73 |
-
| 0.
|
74 |
-
| 0.
|
75 |
-
| 0.
|
76 |
-
| 0.
|
77 |
-
| 0.
|
78 |
-
| 0.
|
79 |
-
| 0.
|
80 |
-
| 0.
|
81 |
-
| 0.
|
82 |
-
| 0.
|
83 |
-
| 0.
|
84 |
-
| 0.
|
85 |
-
| 0.
|
86 |
-
| 0.
|
87 |
-
| 0.
|
88 |
-
| 0.
|
89 |
-
| 0.
|
90 |
-
| 0.
|
91 |
-
| 0.
|
92 |
-
| 0.
|
93 |
-
| 0.
|
94 |
-
| 0.
|
95 |
-
| 0.
|
96 |
-
| 0.
|
97 |
-
| 0.
|
98 |
-
| 0.
|
99 |
|
100 |
|
101 |
### Framework versions
|
102 |
|
103 |
- Transformers 4.28.0
|
104 |
-
- Pytorch 2.1.0+
|
105 |
-
- Datasets 2.
|
106 |
- Tokenizers 0.13.3
|
|
|
1 |
---
|
2 |
license: other
|
3 |
tags:
|
4 |
+
- vision
|
5 |
+
- image-segmentation
|
6 |
- generated_from_trainer
|
7 |
model-index:
|
8 |
- name: segformer-b0-finetuned-segments-toolwear
|
|
|
14 |
|
15 |
# segformer-b0-finetuned-segments-toolwear
|
16 |
|
17 |
+
This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on the HorcruxNo13/toolwear_complete_wear dataset.
|
18 |
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 0.0354
|
20 |
+
- Mean Iou: 0.3022
|
21 |
+
- Mean Accuracy: 0.6045
|
22 |
+
- Overall Accuracy: 0.6045
|
23 |
- Accuracy Unlabeled: nan
|
24 |
- Accuracy Tool: nan
|
25 |
+
- Accuracy Wear: 0.6045
|
26 |
- Iou Unlabeled: 0.0
|
27 |
- Iou Tool: nan
|
28 |
+
- Iou Wear: 0.6045
|
29 |
|
30 |
## Model description
|
31 |
|
|
|
56 |
|
57 |
| Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy Unlabeled | Accuracy Tool | Accuracy Wear | Iou Unlabeled | Iou Tool | Iou Wear |
|
58 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:------------------:|:-------------:|:-------------:|:-------------:|:--------:|:--------:|
|
59 |
+
| 0.8671 | 1.18 | 20 | 0.9263 | 0.4061 | 0.8122 | 0.8122 | nan | nan | 0.8122 | 0.0 | nan | 0.8122 |
|
60 |
+
| 0.5691 | 2.35 | 40 | 0.5998 | 0.2895 | 0.5790 | 0.5790 | nan | nan | 0.5790 | 0.0 | nan | 0.5790 |
|
61 |
+
| 0.4378 | 3.53 | 60 | 0.3948 | 0.3106 | 0.6213 | 0.6213 | nan | nan | 0.6213 | 0.0 | nan | 0.6213 |
|
62 |
+
| 0.3842 | 4.71 | 80 | 0.3190 | 0.2679 | 0.5357 | 0.5357 | nan | nan | 0.5357 | 0.0 | nan | 0.5357 |
|
63 |
+
| 0.3234 | 5.88 | 100 | 0.2883 | 0.3574 | 0.7148 | 0.7148 | nan | nan | 0.7148 | 0.0 | nan | 0.7148 |
|
64 |
+
| 0.2731 | 7.06 | 120 | 0.2392 | 0.3456 | 0.6911 | 0.6911 | nan | nan | 0.6911 | 0.0 | nan | 0.6911 |
|
65 |
+
| 0.2137 | 8.24 | 140 | 0.1850 | 0.1844 | 0.3688 | 0.3688 | nan | nan | 0.3688 | 0.0 | nan | 0.3688 |
|
66 |
+
| 0.1798 | 9.41 | 160 | 0.1692 | 0.2757 | 0.5515 | 0.5515 | nan | nan | 0.5515 | 0.0 | nan | 0.5515 |
|
67 |
+
| 0.1607 | 10.59 | 180 | 0.1338 | 0.2978 | 0.5956 | 0.5956 | nan | nan | 0.5956 | 0.0 | nan | 0.5956 |
|
68 |
+
| 0.1399 | 11.76 | 200 | 0.1218 | 0.2906 | 0.5811 | 0.5811 | nan | nan | 0.5811 | 0.0 | nan | 0.5811 |
|
69 |
+
| 0.1173 | 12.94 | 220 | 0.1030 | 0.2612 | 0.5224 | 0.5224 | nan | nan | 0.5224 | 0.0 | nan | 0.5224 |
|
70 |
+
| 0.0922 | 14.12 | 240 | 0.0976 | 0.2817 | 0.5633 | 0.5633 | nan | nan | 0.5633 | 0.0 | nan | 0.5633 |
|
71 |
+
| 0.081 | 15.29 | 260 | 0.0795 | 0.3154 | 0.6308 | 0.6308 | nan | nan | 0.6308 | 0.0 | nan | 0.6308 |
|
72 |
+
| 0.0852 | 16.47 | 280 | 0.0716 | 0.2188 | 0.4377 | 0.4377 | nan | nan | 0.4377 | 0.0 | nan | 0.4377 |
|
73 |
+
| 0.0709 | 17.65 | 300 | 0.0680 | 0.2691 | 0.5382 | 0.5382 | nan | nan | 0.5382 | 0.0 | nan | 0.5382 |
|
74 |
+
| 0.073 | 18.82 | 320 | 0.0611 | 0.2830 | 0.5660 | 0.5660 | nan | nan | 0.5660 | 0.0 | nan | 0.5660 |
|
75 |
+
| 0.0602 | 20.0 | 340 | 0.0592 | 0.2829 | 0.5657 | 0.5657 | nan | nan | 0.5657 | 0.0 | nan | 0.5657 |
|
76 |
+
| 0.0547 | 21.18 | 360 | 0.0577 | 0.2842 | 0.5684 | 0.5684 | nan | nan | 0.5684 | 0.0 | nan | 0.5684 |
|
77 |
+
| 0.0554 | 22.35 | 380 | 0.0537 | 0.2613 | 0.5226 | 0.5226 | nan | nan | 0.5226 | 0.0 | nan | 0.5226 |
|
78 |
+
| 0.0515 | 23.53 | 400 | 0.0523 | 0.3076 | 0.6152 | 0.6152 | nan | nan | 0.6152 | 0.0 | nan | 0.6152 |
|
79 |
+
| 0.0444 | 24.71 | 420 | 0.0487 | 0.3063 | 0.6126 | 0.6126 | nan | nan | 0.6126 | 0.0 | nan | 0.6126 |
|
80 |
+
| 0.088 | 25.88 | 440 | 0.0467 | 0.3041 | 0.6082 | 0.6082 | nan | nan | 0.6082 | 0.0 | nan | 0.6082 |
|
81 |
+
| 0.0472 | 27.06 | 460 | 0.0437 | 0.2623 | 0.5245 | 0.5245 | nan | nan | 0.5245 | 0.0 | nan | 0.5245 |
|
82 |
+
| 0.0396 | 28.24 | 480 | 0.0474 | 0.3352 | 0.6704 | 0.6704 | nan | nan | 0.6704 | 0.0 | nan | 0.6704 |
|
83 |
+
| 0.0351 | 29.41 | 500 | 0.0436 | 0.3060 | 0.6120 | 0.6120 | nan | nan | 0.6120 | 0.0 | nan | 0.6120 |
|
84 |
+
| 0.0392 | 30.59 | 520 | 0.0428 | 0.2975 | 0.5951 | 0.5951 | nan | nan | 0.5951 | 0.0 | nan | 0.5951 |
|
85 |
+
| 0.0317 | 31.76 | 540 | 0.0431 | 0.3253 | 0.6507 | 0.6507 | nan | nan | 0.6507 | 0.0 | nan | 0.6507 |
|
86 |
+
| 0.0391 | 32.94 | 560 | 0.0404 | 0.2863 | 0.5726 | 0.5726 | nan | nan | 0.5726 | 0.0 | nan | 0.5726 |
|
87 |
+
| 0.0309 | 34.12 | 580 | 0.0408 | 0.3215 | 0.6429 | 0.6429 | nan | nan | 0.6429 | 0.0 | nan | 0.6429 |
|
88 |
+
| 0.0493 | 35.29 | 600 | 0.0381 | 0.2581 | 0.5162 | 0.5162 | nan | nan | 0.5162 | 0.0 | nan | 0.5162 |
|
89 |
+
| 0.0321 | 36.47 | 620 | 0.0376 | 0.3147 | 0.6293 | 0.6293 | nan | nan | 0.6293 | 0.0 | nan | 0.6293 |
|
90 |
+
| 0.0333 | 37.65 | 640 | 0.0372 | 0.3118 | 0.6236 | 0.6236 | nan | nan | 0.6236 | 0.0 | nan | 0.6236 |
|
91 |
+
| 0.0295 | 38.82 | 660 | 0.0362 | 0.3036 | 0.6072 | 0.6072 | nan | nan | 0.6072 | 0.0 | nan | 0.6072 |
|
92 |
+
| 0.0302 | 40.0 | 680 | 0.0365 | 0.3157 | 0.6314 | 0.6314 | nan | nan | 0.6314 | 0.0 | nan | 0.6314 |
|
93 |
+
| 0.0272 | 41.18 | 700 | 0.0367 | 0.3012 | 0.6024 | 0.6024 | nan | nan | 0.6024 | 0.0 | nan | 0.6024 |
|
94 |
+
| 0.0278 | 42.35 | 720 | 0.0353 | 0.2935 | 0.5870 | 0.5870 | nan | nan | 0.5870 | 0.0 | nan | 0.5870 |
|
95 |
+
| 0.0283 | 43.53 | 740 | 0.0353 | 0.2970 | 0.5940 | 0.5940 | nan | nan | 0.5940 | 0.0 | nan | 0.5940 |
|
96 |
+
| 0.0256 | 44.71 | 760 | 0.0355 | 0.3090 | 0.6181 | 0.6181 | nan | nan | 0.6181 | 0.0 | nan | 0.6181 |
|
97 |
+
| 0.0365 | 45.88 | 780 | 0.0358 | 0.3008 | 0.6015 | 0.6015 | nan | nan | 0.6015 | 0.0 | nan | 0.6015 |
|
98 |
+
| 0.025 | 47.06 | 800 | 0.0353 | 0.2965 | 0.5930 | 0.5930 | nan | nan | 0.5930 | 0.0 | nan | 0.5930 |
|
99 |
+
| 0.0299 | 48.24 | 820 | 0.0361 | 0.3109 | 0.6219 | 0.6219 | nan | nan | 0.6219 | 0.0 | nan | 0.6219 |
|
100 |
+
| 0.0239 | 49.41 | 840 | 0.0354 | 0.3022 | 0.6045 | 0.6045 | nan | nan | 0.6045 | 0.0 | nan | 0.6045 |
|
101 |
|
102 |
|
103 |
### Framework versions
|
104 |
|
105 |
- Transformers 4.28.0
|
106 |
+
- Pytorch 2.1.0+cu121
|
107 |
+
- Datasets 2.16.0
|
108 |
- Tokenizers 0.13.3
|