This repository contains the code for PyTorch to CoreML model conversion blog post.

All required libraries collected in the requirements.txt file. To create a new virtual environment and install the requirements, you can use these commands:

$ virtualenv -p python3.7 venv
$ source venv/bin/activate
$ pip install -r requirements.txt

To convert any model from torchvision.models you can try our script by using this command:

$ python3 torch_to_coreml.py --model_name resnet18 --simplify

The --model_name argument should contain the name of any model from torchvision.models

Want to become an expert in AI? AI Courses by OpenCV is a great place to start.