install.md

Installation

Step 1 Create the environment by replacing x by the wanted python version : {3.9}, {3.10} or {3.11}.

conda create -y -n osl-action-spotting python=x

Step 2 Activate the environment

conda activate osl-action-spotting

Step 3 Install the library by replacing x, y and z depending on your cuda version.

If cuda version is {11.2} ~ {11.x} : x, y and z are respectively {11.8}, {110} and {11x}.

If cuda version is {12.x} : x, y and z are respectively {12.1}, {120} and {12x}.

conda install pytorch==2.1.1 torchvision==0.16.1 pytorch-cuda=x -c pytorch -c nvidia
conda install -y cudatoolkit=x -c pytorch
pip install nvidia-dali-cuday
pip install cupy-cudaz
pip install -r requirements.txt
pip install -e .

For example, this is verified for Ubuntu 22.04 python=3.11 cuda version=12.2

conda create -y -n osl python=3.11
conda activate osl
conda install pytorch==2.1.1 torchvision==0.16.1 pytorch-cuda=12.1 -c pytorch -c nvidia
pip install --extra-index-url https://pypi.nvidia.com/ nvidia-dali-cuda120
pip install cupy-cuda12x
pip install -r requirements.txt
pip install "numpy<2.0"
pip install -e .

Note: You might need to downgrade numpy!