Instructions to use lingkai/open-clip with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- OpenCLIP
How to use lingkai/open-clip with OpenCLIP:
import open_clip model, preprocess_train, preprocess_val = open_clip.create_model_and_transforms('hf-hub:lingkai/open-clip') tokenizer = open_clip.get_tokenizer('hf-hub:lingkai/open-clip') - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -26,6 +26,12 @@ The model was trained with the following configuration:
|
|
| 26 |
|
| 27 |
Additional specific performance-enhancing flags enabled during training: `--torchcompile`, `--local-loss`, and `--gather-with-grad`.
|
| 28 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
## Usage
|
| 30 |
|
| 31 |
```python
|
|
|
|
| 26 |
|
| 27 |
Additional specific performance-enhancing flags enabled during training: `--torchcompile`, `--local-loss`, and `--gather-with-grad`.
|
| 28 |
|
| 29 |
+
## Evaluation
|
| 30 |
+
|
| 31 |
+
- **Eval Epoch**: 0
|
| 32 |
+
- **imagenet-zeroshot-val-top1**: 0.6086
|
| 33 |
+
- **imagenet-zeroshot-val-top5**: 0.8632
|
| 34 |
+
|
| 35 |
## Usage
|
| 36 |
|
| 37 |
```python
|