77711be786
* tf.Graph decoder. * Fix conflicts. * Fixed det_input_node() * Added support of non-frozen models. * Cleaned code. * Small fix. * Small corrections. * Error fixes. * Code style. * Code style. * Code style. * Small correction. * Fixed float32 attributes. * Small correction. * Fixed tests. * Fixed errors. * Added statefull partitioned call test. * Import fix. * Code corrections. * BOM test fixed. * Corrected check, added comment. * Added checks. * Supported TF Fraph Iterator in load_by_model(). * Clang format. * Small correction. * Fixed example_input logic, added tests. * Added comment. * Small correction. * Corrected example_input description. * Moved load_by_model test to MO Python API tests. * Minor corrections. * Code corrections. * Small correction. * Clang format. * Fixed tests. * Import change. * Moved GraphIterator to common FE. * Tests refactoring, minor fixes. * Small test correction. * Removed not needed change. * Removed commented code. * Removed not needed change. * Unit tests fix. * Temporarily added debug output. * Test fix. * Applied comments. * Fixed test.
Installation
Installing from PyPi
- Create a virtual environment and activate it, e.g.:
virtualenv -p /usr/bin/python3.7 .env3
source .env3/bin/activate
- Install openvino-dev package, it contains model conversion API:
pip install openvino-dev
This will download all requirements and will install MO in your current virtual environment. If you need only particular frameworks you can specify them manually as optional dependencies in square brackets. E.g. the command below will install dependencies to support ONNX* and TensorFlow2* models:
pip install openvino-dev[onnx,tensorflow2]
To enable support of MxNet* models run:
pip install openvino-dev[mxnet]
To enable support of all frameworks:
pip install openvino-dev[all]
By default, if no frameworks are specified, dependencies to support ONNX* and TensorFlow2* are installed.
Converting models
Setup development environment
If you want to contribute to model conversion API you will need to deploy developer environment. You can do that by following the steps below:
- Create virtual environment and activate it, e.g.:
virtualenv -p /usr/bin/python3.7 .env3
source .env3/bin/activate
- Clone the OpenVINO™ repository and change dir to model-optimizer
git clone https://github.com/openvinotoolkit/openvino
cd openvino/tools/mo/
- Install openvino-mo package for development:
pip install -e .
or run setup.py develop, result will be the same:
python setup.py develop
This will download all requirements and deploy model conversion API for development in your virtual environment: specifically will create *.egg-link into the current directory in your site-packages. As previously noted, you can also manually specify to support only selected frameworks :
pip install -e ".[onnx,tensorflow2]"
How to run unit-tests
- Run tests with:
python -m unittest discover -p "*_test.py" [-s PATH_TO_DIR]
How to capture unit-tests coverage
- Run tests with:
coverage run -m unittest discover -p "*_test.py" [-s PATH_TO_DIR]
- Build html report:
coverage html
How to run code linting
- Run the following command:
pylint openvino/tools/mo/ mo.py