Installation

Docker

We recommend using docker images if docker or other container runtimes e.g. singularity is available on your devices.

We maintain a prebuilt image at dockerhub (cuda version < 11):

yangxue2docker/yx-tf-det:tensorflow1.13.1-cuda10-gpu-py3

Note

For 30xx series graphics cards (cuda version >= 11), please download image from tensorflow-release-notes according to your development environment, e.g. nvcr.io/nvidia/tensorflow:20.11-tf1-py3

Manual configuration

If docker is not available, we provide detailed steps to install the requirements by pip (cuda version < 11):

pip install -r requirements.txt
pip install -v -e .  # or "python setup.py develop"

Or, you can simply install AlphaRotate with the following commands (cuda version < 11):

pip install alpharotate  # Not suitable for dev.

Note

For 30xx series graphics cards (cuda version >= 11), we recommend this blog to install tf1.xx

For 30xx series graphics cards (cuda version >= 11):

cd alpharotate/libs/utils/cython_utils
rm *.so
rm *.c
rm *.cpp
python setup.py build_ext --inplace (or make)

cd alpharotate/libs/utils/
rm *.so
rm *.c
rm *.cpp
python setup.py build_ext --inplace