[Draft] Initial Contents for ORT Mobile Model Zoo

[YUNQIUGUO]

[Draft] Initial Contents for ORT Mobile Model Zoo

• For two nlp cases, models are identical as transformers.js used in their pipeline examples and exported via hf to onnx.
  • https://huggingface.co/docs/transformers.js/main/en/api/pipelines#pipelinesquestionansweringpipeline
  • https://huggingface.co/docs/transformers.js/main/en/api/pipelines#pipelinestextclassificationpipeline
• For two vision cases, models are exported from PyTorch.
• [TODO] For audio, we don't have corresponding pre_post_processing steps implemented yet.

supported_models list from transformers.js: https://github.com/xenova/transformers.js/blob/main/scripts/supported_models.py

Recommend to view the changes/notebooks in branch view: https://github.com/microsoft/onnxruntime-inference-examples/tree/yguo/draft-mobile-model-zoo-notebooks/mobile/model_zoo