Instructions to use visheratin/nllb-siglip-mrl-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- OpenCLIP
How to use visheratin/nllb-siglip-mrl-base with OpenCLIP:
import open_clip model, preprocess_train, preprocess_val = open_clip.create_model_and_transforms('hf-hub:visheratin/nllb-siglip-mrl-base') tokenizer = open_clip.get_tokenizer('hf-hub:visheratin/nllb-siglip-mrl-base') - Notebooks
- Google Colab
- Kaggle
Error loading the model
#2
by AlirezaFa - opened
I get this error when running the provided colab and loading the model:
ValueError: The model class you are passing has a `config_class` attribute that is not consistent with the config class you passed (model has <class 'transformers_modules.visheratin.nllb-siglip-mrl-base.fa0ec582f009620cd758f828792c1ffe509d94ef.nllb_mrl.MatryoshkaNllbClipConfig'> and you passed <class 'transformers_modules.visheratin.nllb-siglip-mrl-base.fa0ec582f009620cd758f828792c1ffe509d94ef.nllb_mrl.MatryoshkaNllbClipConfig'>. Fix one of those so they match!
There is a bug in the latest version of Transformers that will be fixed in this PR. Until then, you can use one of the previous versions of the library or manually patch the latest version using the code from the PR.
I updated the Colab notebook with the working version of Transformers (4.37.0).