search

ltkong218 / IFRNet

IFRNet: Intermediate Feature Refine Network for Efficient Frame Interpolation (CVPR 2022)
MIT License
276 stars 23 forks source link

Training code usage #11

Closed abhishri-medewar closed 2 years ago

abhishri-medewar commented 2 years ago

Hello, Can you provide the details regarding the environment setup which is required to run the training code? I am trying to run it on A10 with 2 GPU's. Will I have to make any changes to run the code?

Thank you in advance!!

ltkong218 commented 2 years ago

I use the basic Anaconda3 python environment, and then install PyTorch, opencv-python and cupy. Python 3.6/3.7 and PyTorch 1.3.0-1.9.1 are verified to be OK. This code does not rely on extra uncommon python packages.

To prepare the training environment, for example: 1) Download Anaconda3-5.3.1-Linux-x86_64.sh and install it on Ubuntu 20.04. 2) Run conda install pytorch==1.9.1 torchvision==0.10.1 torchaudio==0.9.1 cudatoolkit=10.2 -c pytorch to install PyTorch 1.9.1 with CUDA 10.2. 3) Run pip install opencv-python to install opencv python package. 4) Run pip install cupy-cuda102 to install cupy with CUDA 10.2.

You can do some adjustment according to your Python, PyTorch and CUDA requirements. I think there is no need to make any other changes to run the code after you have set up above environment .