The code is developed and tested using 4 or 8 Tesla V100 GPUs.
-
Clone this repo.
$ git clone https://github.com/qpc1611094/FPL.git
-
Install dependencies.
(1) Create a conda environment:
$ conda env create -f FPL.yaml $ conda activate FPL_dist
(2) Install apex 0.1(needs CUDA)
$ cd ./FPL_based_on_CPS/furnace/apex $ python setup.py install --cpp_ext --cuda_ext