Instructions to use Letian2003/unified_vit_v56-vision-decoder with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- OpenCLIP
How to use Letian2003/unified_vit_v56-vision-decoder with OpenCLIP:
import open_clip model, preprocess_train, preprocess_val = open_clip.create_model_and_transforms('hf-hub:Letian2003/unified_vit_v56-vision-decoder') tokenizer = open_clip.get_tokenizer('hf-hub:Letian2003/unified_vit_v56-vision-decoder') - Notebooks
- Google Colab
- Kaggle
| { | |
| "model_cfg": { | |
| "embed_dim": 768, | |
| "vision_cfg": { | |
| "width": 768, | |
| "layers": 12, | |
| "patch_size": 1, | |
| "image_size": 16, | |
| "no_ln_pre": true, | |
| "pool_type": "avg", | |
| "final_ln_after_pool": true, | |
| "norm_kwargs": { | |
| "eps": 1e-06 | |
| }, | |
| "output_tokens": true, | |
| "head_width": 64, | |
| "in_channels": 768 | |
| }, | |
| "text_cfg": { | |
| "context_length": 77, | |
| "vocab_size": 49408, | |
| "width": 512, | |
| "heads": 8, | |
| "layers": 12 | |
| } | |
| } | |
| } |