Code for EQ-Net: Elastic Quantization Neural Networks accepted by ICCV2023
We give a config sample under the configs folder.
Train Supernet(Multi-GPU only)
python train_supernet.py --cfg cfg_path
Download the Trained Supermodel Google-Drive / BaiduDisk
Please fill in the trained Supernet model path in the configuration file, then run
python eval_supernet.py --cfg cfg_path
Part of the code references pytorch-BigNAS and LSQplus. Please ref to their reposity for more details.