Hugging Face's logo

nandyc
/
swin-tiny-patch4-window7-224-finetuned_ASL_Isolated_Swin_dataset2

Image Classification
Transformers
PyTorch
swin
Generated from Trainer
Inference Endpoints
Model card Files Community
nandyc commited on
Commit
00b156e
•
1 Parent(s): 9e1182a

Model save

Browse files
Files changed (3) hide show
  1. README.md +22 -51
  2. pytorch_model.bin +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -2,7 +2,6 @@
2
  license: apache-2.0
3
  base_model: microsoft/swin-tiny-patch4-window7-224
4
  tags:
5
- - image-classification
6
  - generated_from_trainer
7
  metrics:
8
  - accuracy
@@ -16,10 +15,10 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  # swin-tiny-patch4-window7-224-finetuned_ASL_Isolated_Swin_dataset2
18
 
19
- This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the ASL_Isolated_Swin_dataset dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.1246
22
- - Accuracy: 0.9885
23
 
24
  ## Model description
25
 
@@ -44,58 +43,30 @@ The following hyperparameters were used during training:
44
  - seed: 42
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
47
- - num_epochs: 50
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
52
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
53
- | 1.4683 | 1.09 | 100 | 1.1745 | 0.5692 |
54
- | 1.0278 | 2.17 | 200 | 0.5680 | 0.8385 |
55
- | 0.6458 | 3.26 | 300 | 0.2887 | 0.8808 |
56
- | 0.6043 | 4.35 | 400 | 0.2451 | 0.9385 |
57
- | 0.5372 | 5.43 | 500 | 0.2226 | 0.9231 |
58
- | 0.4619 | 6.52 | 600 | 0.3455 | 0.9115 |
59
- | 0.5265 | 7.61 | 700 | 0.3178 | 0.9154 |
60
- | 0.284 | 8.7 | 800 | 0.4421 | 0.8885 |
61
- | 0.4421 | 9.78 | 900 | 0.2261 | 0.9308 |
62
- | 0.3366 | 10.87 | 1000 | 0.2085 | 0.9423 |
63
- | 0.4313 | 11.96 | 1100 | 0.1938 | 0.9654 |
64
- | 0.4708 | 13.04 | 1200 | 0.0983 | 0.9615 |
65
- | 0.5222 | 14.13 | 1300 | 0.2471 | 0.9385 |
66
- | 0.4582 | 15.22 | 1400 | 0.0927 | 0.9808 |
67
- | 0.3018 | 16.3 | 1500 | 0.2100 | 0.9385 |
68
- | 0.3666 | 17.39 | 1600 | 0.2686 | 0.95 |
69
- | 0.2042 | 18.48 | 1700 | 0.2468 | 0.95 |
70
- | 0.4278 | 19.57 | 1800 | 0.1992 | 0.9692 |
71
- | 0.2769 | 20.65 | 1900 | 0.1939 | 0.95 |
72
- | 0.2471 | 21.74 | 2000 | 0.1791 | 0.95 |
73
- | 0.3054 | 22.83 | 2100 | 0.1815 | 0.9615 |
74
- | 0.4803 | 23.91 | 2200 | 0.2399 | 0.95 |
75
- | 0.218 | 25.0 | 2300 | 0.2819 | 0.9462 |
76
- | 0.1858 | 26.09 | 2400 | 0.2068 | 0.9769 |
77
- | 0.214 | 27.17 | 2500 | 0.2683 | 0.9462 |
78
- | 0.2532 | 28.26 | 2600 | 0.1364 | 0.9615 |
79
- | 0.2021 | 29.35 | 2700 | 0.1507 | 0.9692 |
80
- | 0.1794 | 30.43 | 2800 | 0.1450 | 0.9692 |
81
- | 0.1263 | 31.52 | 2900 | 0.1562 | 0.9692 |
82
- | 0.1298 | 32.61 | 3000 | 0.1078 | 0.9808 |
83
- | 0.1883 | 33.7 | 3100 | 0.0763 | 0.9846 |
84
- | 0.1773 | 34.78 | 3200 | 0.0770 | 0.9846 |
85
- | 0.1491 | 35.87 | 3300 | 0.0956 | 0.9846 |
86
- | 0.1866 | 36.96 | 3400 | 0.1162 | 0.9846 |
87
- | 0.1269 | 38.04 | 3500 | 0.1366 | 0.9769 |
88
- | 0.1498 | 39.13 | 3600 | 0.1246 | 0.9885 |
89
- | 0.1981 | 40.22 | 3700 | 0.0920 | 0.9808 |
90
- | 0.1421 | 41.3 | 3800 | 0.0983 | 0.9808 |
91
- | 0.1936 | 42.39 | 3900 | 0.1123 | 0.9769 |
92
- | 0.097 | 43.48 | 4000 | 0.1506 | 0.9769 |
93
- | 0.0791 | 44.57 | 4100 | 0.1443 | 0.9769 |
94
- | 0.0935 | 45.65 | 4200 | 0.1689 | 0.9808 |
95
- | 0.1061 | 46.74 | 4300 | 0.1379 | 0.9808 |
96
- | 0.2041 | 47.83 | 4400 | 0.1391 | 0.9769 |
97
- | 0.0262 | 48.91 | 4500 | 0.1360 | 0.9769 |
98
- | 0.1274 | 50.0 | 4600 | 0.1370 | 0.9769 |
99
 
100
 
101
  ### Framework versions
 
2
  license: apache-2.0
3
  base_model: microsoft/swin-tiny-patch4-window7-224
4
  tags:
 
5
  - generated_from_trainer
6
  metrics:
7
  - accuracy
 
15
 
16
  # swin-tiny-patch4-window7-224-finetuned_ASL_Isolated_Swin_dataset2
17
 
18
+ This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 0.0504
21
+ - Accuracy: 0.9769
22
 
23
  ## Model description
24
 
 
43
  - seed: 42
44
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
  - lr_scheduler_type: linear
46
+ - num_epochs: 20
47
 
48
  ### Training results
49
 
50
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
51
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
52
+ | 1.6758 | 1.09 | 100 | 1.3206 | 0.5769 |
53
+ | 1.0717 | 2.17 | 200 | 0.6482 | 0.8154 |
54
+ | 0.627 | 3.26 | 300 | 0.4486 | 0.8654 |
55
+ | 0.5397 | 4.35 | 400 | 0.2978 | 0.8923 |
56
+ | 0.537 | 5.43 | 500 | 0.1513 | 0.9423 |
57
+ | 0.3766 | 6.52 | 600 | 0.4737 | 0.8846 |
58
+ | 0.3994 | 7.61 | 700 | 0.3060 | 0.9115 |
59
+ | 0.2139 | 8.7 | 800 | 0.1345 | 0.9577 |
60
+ | 0.2995 | 9.78 | 900 | 0.1558 | 0.95 |
61
+ | 0.2835 | 10.87 | 1000 | 0.0943 | 0.9731 |
62
+ | 0.3089 | 11.96 | 1100 | 0.0913 | 0.9577 |
63
+ | 0.3632 | 13.04 | 1200 | 0.0888 | 0.9692 |
64
+ | 0.327 | 14.13 | 1300 | 0.1038 | 0.9808 |
65
+ | 0.313 | 15.22 | 1400 | 0.0976 | 0.9731 |
66
+ | 0.1752 | 16.3 | 1500 | 0.0504 | 0.9808 |
67
+ | 0.2397 | 17.39 | 1600 | 0.0612 | 0.9808 |
68
+ | 0.1348 | 18.48 | 1700 | 0.0558 | 0.9846 |
69
+ | 0.2842 | 19.57 | 1800 | 0.0504 | 0.9769 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
70
 
71
 
72
  ### Framework versions
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:44f2e2bafe3bfebb97b55f91f3d6bc61e877f5da8c51daaf7470ac42815e1a6f
3
  size 110469102
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a538e283c914ab0e84c69183b09e90ba99a1ee1a954fb18649b7196487d21a82
3
  size 110469102
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:aab1838563af5a2f1d5a88a860374890aa5db23feb5f96f4b01360d855f6dfb9
3
  size 4600
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c204df3758aed9eb75e651347c36195f5ebfe556457339e4f8d1b5c5874134dd
3
  size 4600