Initial release of infernet-ml, a library to build Web3 machine learning workflows.
Features
- Workflow classes for running inference:
- `HFInferenceClientWorkflow`: Wrapper around [Huggingface's Inference Client library](https://huggingface.co/docs/huggingface_hub/en/package_reference/inference_client)
- `ONNXInferenceWorkflow`: ONNX model suport
- `TorchInferenceWorkflow`: PyTorch model support
- `TTSInferenceWorkflow`: Base class for text-to-speech model support
- `TGIClientInferenceWorkflow`: [Huggingface's Text Generation Interface](https://huggingface.co/docs/text-generation-inference/en/index) server inference support
- `BarkHFInferenceWorkflow`: Suno's Bark model support
- Utility for model retrieval from different sources, beginning with:
- Hugginface Hub
- Arweave