Hugging Face's logo

FatihC
/
swin-tiny-patch4-window7-224-finetuned-eurosat-watermark

Image Classification
Transformers
PyTorch
TensorBoard
swin
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
3
FatihC commited on
Commit
8ef239d
•
1 Parent(s): d15f4ae

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +116 -0
README.md ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - imagefolder
7
+ metrics:
8
+ - accuracy
9
+ model-index:
10
+ - name: swin-tiny-patch4-window7-224-finetuned-eurosat
11
+ results:
12
+ - task:
13
+ name: Image Classification
14
+ type: image-classification
15
+ dataset:
16
+ name: imagefolder
17
+ type: imagefolder
18
+ config: images
19
+ split: train
20
+ args: images
21
+ metrics:
22
+ - name: Accuracy
23
+ type: accuracy
24
+ value: 0.953125
25
+ ---
26
+
27
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
28
+ should probably proofread and complete it, then remove this comment. -->
29
+
30
+ # swin-tiny-patch4-window7-224-finetuned-eurosat
31
+
32
+ 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 imagefolder dataset.
33
+ It achieves the following results on the evaluation set:
34
+ - Loss: 0.1379
35
+ - Accuracy: 0.9531
36
+
37
+ ## Model description
38
+
39
+ More information needed
40
+
41
+ ## Intended uses & limitations
42
+
43
+ More information needed
44
+
45
+ ## Training and evaluation data
46
+
47
+ More information needed
48
+
49
+ ## Training procedure
50
+
51
+ ### Training hyperparameters
52
+
53
+ The following hyperparameters were used during training:
54
+ - learning_rate: 5e-05
55
+ - train_batch_size: 32
56
+ - eval_batch_size: 32
57
+ - seed: 42
58
+ - gradient_accumulation_steps: 4
59
+ - total_train_batch_size: 128
60
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
+ - lr_scheduler_type: linear
62
+ - lr_scheduler_warmup_ratio: 0.1
63
+ - num_epochs: 40
64
+
65
+ ### Training results
66
+
67
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
68
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
69
+ | No log | 1.0 | 4 | 0.4862 | 0.8516 |
70
+ | No log | 2.0 | 8 | 0.4103 | 0.8828 |
71
+ | 0.4518 | 3.0 | 12 | 0.3210 | 0.8984 |
72
+ | 0.4518 | 4.0 | 16 | 0.2053 | 0.9375 |
73
+ | 0.2909 | 5.0 | 20 | 0.1675 | 0.9453 |
74
+ | 0.2909 | 6.0 | 24 | 0.1439 | 0.9531 |
75
+ | 0.2909 | 7.0 | 28 | 0.1448 | 0.9297 |
76
+ | 0.1492 | 8.0 | 32 | 0.1798 | 0.9531 |
77
+ | 0.1492 | 9.0 | 36 | 0.1360 | 0.9453 |
78
+ | 0.1161 | 10.0 | 40 | 0.1670 | 0.9531 |
79
+ | 0.1161 | 11.0 | 44 | 0.1637 | 0.9531 |
80
+ | 0.1161 | 12.0 | 48 | 0.1298 | 0.9531 |
81
+ | 0.1053 | 13.0 | 52 | 0.1162 | 0.9531 |
82
+ | 0.1053 | 14.0 | 56 | 0.1353 | 0.9531 |
83
+ | 0.0839 | 15.0 | 60 | 0.1211 | 0.9609 |
84
+ | 0.0839 | 16.0 | 64 | 0.1113 | 0.9609 |
85
+ | 0.0839 | 17.0 | 68 | 0.1145 | 0.9609 |
86
+ | 0.0689 | 18.0 | 72 | 0.1239 | 0.9531 |
87
+ | 0.0689 | 19.0 | 76 | 0.1280 | 0.9531 |
88
+ | 0.0581 | 20.0 | 80 | 0.1533 | 0.9531 |
89
+ | 0.0581 | 21.0 | 84 | 0.1323 | 0.9609 |
90
+ | 0.0581 | 22.0 | 88 | 0.1327 | 0.9531 |
91
+ | 0.0545 | 23.0 | 92 | 0.1529 | 0.9531 |
92
+ | 0.0545 | 24.0 | 96 | 0.1357 | 0.9531 |
93
+ | 0.046 | 25.0 | 100 | 0.1333 | 0.9531 |
94
+ | 0.046 | 26.0 | 104 | 0.1466 | 0.9531 |
95
+ | 0.046 | 27.0 | 108 | 0.1300 | 0.9531 |
96
+ | 0.0421 | 28.0 | 112 | 0.1077 | 0.9609 |
97
+ | 0.0421 | 29.0 | 116 | 0.0985 | 0.9609 |
98
+ | 0.0371 | 30.0 | 120 | 0.1186 | 0.9531 |
99
+ | 0.0371 | 31.0 | 124 | 0.1123 | 0.9531 |
100
+ | 0.0371 | 32.0 | 128 | 0.1144 | 0.9531 |
101
+ | 0.0348 | 33.0 | 132 | 0.1276 | 0.9531 |
102
+ | 0.0348 | 34.0 | 136 | 0.1488 | 0.9531 |
103
+ | 0.0211 | 35.0 | 140 | 0.1560 | 0.9531 |
104
+ | 0.0211 | 36.0 | 144 | 0.1477 | 0.9531 |
105
+ | 0.0211 | 37.0 | 148 | 0.1488 | 0.9531 |
106
+ | 0.0274 | 38.0 | 152 | 0.1467 | 0.9531 |
107
+ | 0.0274 | 39.0 | 156 | 0.1401 | 0.9531 |
108
+ | 0.0259 | 40.0 | 160 | 0.1379 | 0.9531 |
109
+
110
+
111
+ ### Framework versions
112
+
113
+ - Transformers 4.28.1
114
+ - Pytorch 2.0.0+cu118
115
+ - Datasets 2.11.0
116
+ - Tokenizers 0.13.3