File size: 867 Bytes
7b11ced
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
---
base_model:
- openai/clip-vit-base-patch32
datasets:
- tanganke/gtsrb
metrics:
- accuracy
---

# Model Card

## Model Details

- Architecture: ViT-Base with patch size 32
- Training Data: GTSRB dataset

## Training Details

  Adam Optimizer with a constant learning rate 1e-5 for 4000 steps training (batch_size=32).
  Only the vision encoder is fine-tuned.

## Evaluation Results

- pre-trained: 0.3042911887168884
- fine-tuned: 0.9881029725074768

## Usage

load vision model

```python
from transformers import CLIPVisionModel

vision_model = CLIPVisionModel.from_pretrained('tanganke/clip-vit-base-patch32_gtsrb')
```

substitute the vision encoder of clip

```python
from transformers import CLIPModel

clip_model = CLIPModel.from_pretrained("openai/clip-vit-base-patch32")
clip_model.vision_model.load_state_dict(vision_model.vision_model.state_dict())
```