文件名称:haq:[CVPR 2019,口头] HAQ
文件大小:54KB
文件格式:ZIP
更新时间:2024-05-28 09:10:47
quantization automl mixed-precision efficient-model Python
HAQ:具有混合精度的硬件感知自动量化 介绍 此存储库包含纸质PyTorch实施(CVPR2019,口头) @inproceedings{haq, author = {Wang, Kuan and Liu, Zhijian and Lin, Yujun and Lin, Ji and Han, Song}, title = {HAQ: Hardware-Aware Automated Quantization With Mixed Precision}, booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)}, year = {2019} } 其他与自动化模型设计有关的论文: AMC:用于移动设备上模型压缩和加速的 ( ) ProxylessNAS:在目标任务和硬件上进行直接神经
【文件预览】:
haq-master
----run()
--------run_linear_quantize_eval.sh(390B)
--------run_kmeans_quantize_eval.sh(313B)
--------run_linear_quantize_finetune.sh(521B)
--------run_kmeans_quantize_search.sh(379B)
--------run_pretrain.sh(486B)
--------run_linear_quantize_search.sh(568B)
--------run_kmeans_quantize_finetune.sh(317B)
--------setup.sh(179B)
----models()
--------mobilenetv3.py(8KB)
--------mobilenetv2.py(7KB)
--------__init__.py(112B)
--------mobilenet.py(4KB)
----lib()
--------utils()
--------env()
--------__init__.py(0B)
--------simulator()
--------rl()
----pretrain.py(15KB)
----requirements.txt(126B)
----rl_quantize.py(13KB)
----finetune.py(17KB)
----LICENSE(1KB)
----README.md(6KB)