# apex

**Repository Path**: njcky/apex

## Basic Information

- **Project Name**: apex
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: BSD-3-Clause
- **Default Branch**: torch17cu102
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2021-04-29
- **Last Updated**: 2021-07-14

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

CUDA 10.2 conda pytorch 1.7 环境下
```
https://github.com/NVIDIA/apex/issues/1091#issuecomment-826578213
git reset --hard a651e2c24ecf97cbf367fd3f330df36760e1c597
pip install -v --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" .
```
否则使用以下命令
```
pip install -v --no-cache-dir .` (without CUDA/C++ extensions) is more likely to work.  
```
但不会编译一些融合op，加速效果不如前者