Neural Networks23(CCF-B) - Long-range zero-shot generative deep network quantization paper
conda create -n mindspore python=3.8
pip install mindspore
pip install -r requirement.txt
The pre-trained models and corresponding logs can be downloaded here
Please make sure the "qw" and "qa" in *.hocon, *.hocon, "--model_name" and "--model_path" are correct.
For cifar10
python test.py --model_name resnet20_cifar10 --model_path path_to_pre-trained model --conf_path cifar10_resnet20.hocon
or
python test.py --model_name resnet20_cifar100 --model_path path_to_pre-trained model --conf_path cifar100_resnet20.hocon
For ImageNet
python test.py --model_name resnet18/mobilenet_w1/mobilenetv2_w1 --model_path path_to_pre-trained model --conf_path imagenet.hocon
Results of pre-trained models are shown below:
| Model | Bit-width | Dataset | Top-1 Acc. |
|---|---|---|---|
| resnet18 | W4A4 | ImageNet | 66.47% |
| resnet18 | W5A5 | ImageNet | 69.94% |
| mobilenetv1 | W4A4 | ImageNet | 51.36% |
| mobilenetv1 | W5A5 | ImageNet | 68.17% |
| mobilenetv2 | W4A4 | ImageNet | 65.10% |
| mobilenetv2 | W5A5 | ImageNet | 71.28% |
| resnet-20 | W3A3 | cifar10 | 77.07% |
| resnet-20 | W4A4 | cifar10 | 91.49% |
| resnet-20 | W3A3 | cifar100 | 64.98% |
| resnet-20 | W4A4 | cifar100 | 48.25% |