This notebook shows how to convert model from original framework format to the OpenVINO IR. It describes Python conversion API and Model Optimizer command-line tool. It provides examples of Hugging Face and Pytorch models conversion to OpenVINO IR.
The OpenVINO IR conversion tutorial consists of the following content:
- OpenVINO IR format
- IR preparation with Python conversion API and Model Optimizer command-line tool
- Fetching example models
- Basic conversion
- Model conversion parameters
- Setting Input Shapes
- Cutting Off Parts of a Model
- Embedding Preprocessing Computation
- Specifying Layout
- Changing Model Layout
- Specifying Mean and Scale Values
- Reversing Input Channels
- Compressing a Model to FP16
- Convert Models Represented as Python Objects
This is a self-contained example that relies solely on its own code.
We recommend running the notebook in a virtual environment. You only need a Jupyter server to start. For details, please refer to Installation Guide.