Showcase performance of PyTorch Image Models (Timm) with OpenVINO

My proposal aims to optimize PyTorch Image Models (Timm) using OpenVINO and NNCF libraries. The problem at hand is that these models are computationally intensive and require a lot of resources, making them challenging to deploy in resource-limited environments. To solve this problem, I plan to implement optimization techniques such as quantization, pruning, and compression on a selected subset of models from the Timm library. I will evaluate the performance of these optimized models on different hardware platforms and compare them with the baseline models. The deliverables for this project include optimized Timm models, a tutorial notebook detailing the setup and usage of OpenVINO and NNCF libraries, and documentation and tests for the code developed during the project. If my proposal is accepted, I am excited to work on this project and contribute to the community by making these models more accessible and efficient to deploy.