The load_in_4bit and load_in_8bit arguments are deprecated and will be removed in the future versions. Please, pass a BitsAndBytesConfig object in quantization_config argument instead.

pytorch to TF lite

https://colab.research.google.com/github/joshuacwnewton/ONNX-to-TFLite/blob/master/onnx_to_tflite.ipynb#scrollTo=y4h6nfXPxs7g

https://launchx.netspresso.ai/main