compares different pretrained object classification with per-layer and per-channel quantization using pytorch
Steps to work with google colab
1. Clone the repository
2. Add the path to sys.path so that libraries can be traversed correctly
3. create folder in google colab /content/data/
4. import the compare_models.py like from compare_models import compare_model
5. This package supports AlexNet,MobileNetv2, resnet50, resnet18 models only
6. Run the following commands
6. list = ["AlexNet","MobileNetv2", "resnet50", "resnet18"] //creating a list
7. statistics = compare_model(list) //passing the list as argument such that these models will be compared