Instructions to use apple/DFN5B-CLIP-ViT-H-14 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- OpenCLIP
How to use apple/DFN5B-CLIP-ViT-H-14 with OpenCLIP:
import open_clip model, preprocess_train, preprocess_val = open_clip.create_model_and_transforms('hf-hub:apple/DFN5B-CLIP-ViT-H-14') tokenizer = open_clip.get_tokenizer('hf-hub:apple/DFN5B-CLIP-ViT-H-14') - Notebooks
- Google Colab
- Kaggle
Conversion to ONNX
#4
by sushilkhadka - opened
Does anyone have any idea on converting this model from open_clip to onnx format. I tried following official pytorch documentation, and I'm able to get the onnx file but the catch here is it's not working for 'CUDAExecutionProvider', it's saying ArgMax Not Implemented (13), so I lowered the opset to 11 from 14, now it's not even producing onnx format, saying some other layer not implement. Has anyone solved this issue?