This repository contains Google Colab Notebooks for simple ML/DL Models experiments. The primary target is Model Inference.
- These script can be used as reference for creating Operator/Layer level models for unit level testing of Inference Software
- create_onnx_model.ipynb - Export ONNX model using Pytorch
- define_onnx_model.ipynb - Define ONNX model using onnx.helper
- create_tflite_model.ipynb
- These scripts can be used as reference for developing inference software for Pretrained DL models.
- infer_caffe.ipynb
- infer_ort.ipynb
- infer_tflite.ipynb
- These scripts can be used as reference for saving onnx/tflite models from public domain model hubs.
- save_keras_model.ipynb
- save_timm_model.ipynb
- save_torchvision_model.ipynb