Hugging Face's logo

sungile
/
custom-object-test3

Image Segmentation
Transformers
Safetensors
segformer
vision
Generated from Trainer
Inference Endpoints
Model card Files Community
sungile commited on
Commit
bc049cd
·
verified ·
1 Parent(s): f4e08df

End of training

Browse files
Files changed (4) hide show
  1. README.md +15 -15
  2. config.json +1 -1
  3. model.safetensors +1 -1
  4. training_args.bin +1 -1
README.md CHANGED
@@ -18,16 +18,16 @@ 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 sungile/custom-object-masking3 dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.3334
22
- - Mean Iou: 0.5000
23
- - Mean Accuracy: 1.0000
24
- - Overall Accuracy: 1.0000
25
  - Accuracy Unknown: nan
26
- - Accuracy Background: 1.0000
27
  - Accuracy Object: nan
28
  - Iou Unknown: nan
29
- - Iou Background: 1.0000
30
- - Iou Object: 0.0
31
 
32
  ## Model description
33
 
@@ -58,14 +58,14 @@ The following hyperparameters were used during training:
58
 
59
  | Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy Unknown | Accuracy Background | Accuracy Object | Iou Unknown | Iou Background | Iou Object |
60
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:----------------:|:-------------------:|:---------------:|:-----------:|:--------------:|:----------:|
61
- | 0.8217 | 0.25 | 20 | 0.8519 | 0.3321 | 0.9964 | 0.9964 | nan | 0.9964 | nan | 0.0 | 0.9964 | 0.0 |
62
- | 0.6554 | 0.5 | 40 | 0.5881 | 0.3312 | 0.9937 | 0.9937 | nan | 0.9937 | nan | 0.0 | 0.9937 | 0.0 |
63
- | 0.5488 | 0.75 | 60 | 0.4798 | 0.3316 | 0.9948 | 0.9948 | nan | 0.9948 | nan | 0.0 | 0.9948 | 0.0 |
64
- | 0.469 | 1.0 | 80 | 0.4007 | 0.3326 | 0.9979 | 0.9979 | nan | 0.9979 | nan | 0.0 | 0.9979 | 0.0 |
65
- | 0.438 | 1.25 | 100 | 0.3723 | 0.5000 | 1.0000 | 1.0000 | nan | 1.0000 | nan | nan | 1.0000 | 0.0 |
66
- | 0.3821 | 1.5 | 120 | 0.3781 | 0.3324 | 0.9971 | 0.9971 | nan | 0.9971 | nan | 0.0 | 0.9971 | 0.0 |
67
- | 0.3472 | 1.75 | 140 | 0.3435 | 0.5000 | 1.0000 | 1.0000 | nan | 1.0000 | nan | nan | 1.0000 | 0.0 |
68
- | 0.3473 | 2.0 | 160 | 0.3334 | 0.5000 | 1.0000 | 1.0000 | nan | 1.0000 | nan | nan | 1.0000 | 0.0 |
69
 
70
 
71
  ### Framework versions
 
18
 
19
  This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on the sungile/custom-object-masking3 dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.6728
22
+ - Mean Iou: nan
23
+ - Mean Accuracy: nan
24
+ - Overall Accuracy: nan
25
  - Accuracy Unknown: nan
26
+ - Accuracy Background: nan
27
  - Accuracy Object: nan
28
  - Iou Unknown: nan
29
+ - Iou Background: nan
30
+ - Iou Object: nan
31
 
32
  ## Model description
33
 
 
58
 
59
  | Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy Unknown | Accuracy Background | Accuracy Object | Iou Unknown | Iou Background | Iou Object |
60
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:----------------:|:-------------------:|:---------------:|:-----------:|:--------------:|:----------:|
61
+ | 1.4981 | 0.25 | 20 | 1.5526 | nan | nan | nan | nan | nan | nan | nan | nan | nan |
62
+ | 1.2809 | 0.5 | 40 | 1.4961 | nan | nan | nan | nan | nan | nan | nan | nan | nan |
63
+ | 1.025 | 0.75 | 60 | 1.3031 | nan | nan | nan | nan | nan | nan | nan | nan | nan |
64
+ | 0.9376 | 1.0 | 80 | 0.9104 | nan | nan | nan | nan | nan | nan | nan | nan | nan |
65
+ | 0.8413 | 1.25 | 100 | 0.8012 | nan | nan | nan | nan | nan | nan | nan | nan | nan |
66
+ | 0.7648 | 1.5 | 120 | 0.7258 | nan | nan | nan | nan | nan | nan | nan | nan | nan |
67
+ | 0.7118 | 1.75 | 140 | 0.7015 | nan | nan | nan | nan | nan | nan | nan | nan | nan |
68
+ | 0.6699 | 2.0 | 160 | 0.6728 | nan | nan | nan | nan | nan | nan | nan | nan | nan |
69
 
70
 
71
  ### Framework versions
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "nvidia/mit-b0",
3
  "architectures": [
4
  "SegformerForSemanticSegmentation"
5
  ],
 
1
  {
2
+ "_name_or_path": "sungile/custom-object-test3",
3
  "architectures": [
4
  "SegformerForSemanticSegmentation"
5
  ],
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:28d358aae961b04eee361760c389ada87057b18bdf6719bdcce78e686f8d3b98
3
  size 14885804
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:785276863b0089b0d88c3373cd2af808a23765f3578e0d081ae3a503be86f15f
3
  size 14885804
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ebb801b91b507b138c9f97fdf24ef82622ed740ed8a24ed6c849a891ed584b8b
3
  size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d07fb4280457b422679cf0abf5373f25e113121a24e7eae47ddfc9baa998f71e
3
  size 5304