mrm8488 commited on
Commit
e047f9d
1 Parent(s): f3b6530

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -6
README.md CHANGED
@@ -32,6 +32,12 @@ It achieves the following results on the evaluation set:
32
  - Loss: 0.0549
33
  - Accuracy: 0.9805
34
 
 
 
 
 
 
 
35
  ## Model description
36
 
37
  ConvNeXT is a pure convolutional model (ConvNet), inspired by the design of Vision Transformers, that claims to outperform them. The authors started from a ResNet and "modernized" its design by taking the Swin Transformer as inspiration.
@@ -72,12 +78,6 @@ The following hyperparameters were used during training:
72
  | 0.1435 | 3.0 | 2154 | 0.0549 | 0.9805 |
73
 
74
 
75
- ### Drag and drop the following pics in the left widget to test the model
76
-
77
- ![image1](https://huggingface.co/mrm8488/convnext-tiny-finetuned-eurosat/resolve/main/test1.jpg)
78
- ![image2](https://huggingface.co/mrm8488/convnext-tiny-finetuned-eurosat/resolve/main/test2.jpg)
79
-
80
-
81
  ### Framework versions
82
 
83
  - Transformers 4.18.0
32
  - Loss: 0.0549
33
  - Accuracy: 0.9805
34
 
35
+ #### Drag and drop the following pics in the right widget to test the model
36
+
37
+ ![image1](https://huggingface.co/mrm8488/convnext-tiny-finetuned-eurosat/resolve/main/test1.jpg)
38
+ ![image2](https://huggingface.co/mrm8488/convnext-tiny-finetuned-eurosat/resolve/main/test2.jpg)
39
+
40
+
41
  ## Model description
42
 
43
  ConvNeXT is a pure convolutional model (ConvNet), inspired by the design of Vision Transformers, that claims to outperform them. The authors started from a ResNet and "modernized" its design by taking the Swin Transformer as inspiration.
78
  | 0.1435 | 3.0 | 2154 | 0.0549 | 0.9805 |
79
 
80
 
 
 
 
 
 
 
81
  ### Framework versions
82
 
83
  - Transformers 4.18.0