This repo is used to research large-scale image classification models for embedded systems. For this purpose, the repo contains (re)implementations of various classification models and scripts for training/evaluating/converting.

The following frameworks are used:

• MXNet/Gluon (info),
• PyTorch (info),
• Chainer (info),
• Keras with MXNet backend (info),
• TensorFlow (info).

For each supported framework, there is a PIP-package containing pure models without auxiliary scripts. List of packages:

• gluoncv2 for Gluon,
• pytorchcv for PyTorch,
• chainercv2 for Chainer,
• kerascv for Keras-MXNet,
• tensorflowcv for TensorFlow.

Currently, models are mostly implemented on Gluon and then ported to other frameworks. Some models are pretrained on ImageNet-1K and CIFAR-10/100 datasets. All pretrained weights are loaded automatically during use. See examples of such automatic loading of weights in the corresponding sections of the documentation dedicated to a particular package:

• Gluon models,
• PyTorch models,
• Chainer models,
• Keras models,
• TensorFlow models.

To use training/evaluating scripts as well as all models, you need to clone the repository and install dependencies:

git clone git@github.com:osmr/imgclsmob.git
pip install -r requirements.txt

Some remarks:

• Repo is an author repository, if it exists.
• A, B, and C means the implementation of a model for ImageNet-1K, CIFAR-10, and CIFAR-100, respectively.
• A+, B+, and C+ means having a pre-trained model for ImageNet-1K, CIFAR-10, and CIFAR-100, respectively.