facebookresearch/maskrcnn-benchmark

Half/mixed-precision is not faster

Open

#1,062 创建于 2019年8月27日

在 GitHub 查看
 (3 评论) (0 反应) (0 负责人)Python (9,161 star) (2,574 fork)batch import
help wanted

描述

❓ Questions and Help

As written in https://github.com/facebookresearch/maskrcnn-benchmark/issues/807#issuecomment-500112612 , DTYPE "float16" does not make training faster( and takes the same amount of memory) when tested with e2e_mask_rcnn_R_50_FPN_1x.yaml. Is this normal or am I doing something wrong? I also want to know if I can make inference faster by FP16(I tried model.half() in test_net, but doesn't work).
My environment is RTX Titan, cudnn 7.6, cuda10.0, driver version Driver Version: 418.67, pytorch-1.2.0, ubuntu 18.04.

贡献者指南

技术栈
pythonpytorch
领域
machine learningperformance
议题类型
performance
难度面向新贡献者的预计实现难度，1 表示很小改动，5 表示专家级工作。
3
预计时间有经验贡献者完成调查、实现、测试并准备 pull request 的粗略时间范围。
1-3 hours
活动状态议题当前的可参与程度：新鲜、活跃、陈旧、阻塞或等待维护者输入。
stale
清晰度议题是否清楚说明期望改动、验收标准和下一步。
needs investigation
前置要求
PyTorch basicsmixed precision concepts
新手友好度1-100 的估计分数，表示该议题对首次贡献者的友好程度。
30
研究方向
The issue is a user question about why half precision is not faster in maskrcnn benchmark. To help, one should investigate the use of automatic mixed precision (AMP) in PyTorch, check if the model and training script are properly configured for FP16 (e.g., using torch.cuda.amp or apex), and verify that the GPU supports fast FP16 operations (e.g., RTX Titan). Additionally, check for any known issues in the repository related to mixed precision, such as the linked comment in issue #807. The user also tried model.half() for inference without success, so explore proper FP16 inference workflows.