Skip to content

Latest commit

 

History

History
50 lines (39 loc) · 1.37 KB

INSTALL.md

File metadata and controls

50 lines (39 loc) · 1.37 KB
Raw

We provide instructions to install the required dependencies.

Requirements:

python>=3.7 pytorch==1.7.1 and cuda=10.2 (should work with pytorch >=1.7.1 as well but not tested)

  1. Clone the repo:

    git clone https://github.com/TheShadow29/Video-QAP.git
cd VidSitu
export ROOT=$(pwd)

  2. To use the same environment you can use conda and the environment file vidqap_env.yml file provided. Please refer to Miniconda for details on installing conda.

    MINICONDA_ROOT=[to your Miniconda/Anaconda root directory]
conda env create -f vidqap_env.yml --prefix $MINICONDA_ROOT/envs/vidqap_pyt
conda activate vidqap_pyt

  3. Install fairseq from my fork. Essentially, we need v0.9 of fairseq but one small change is needed to make it work with Pytorch 1.7

    cd $ROOT
git clone https://github.com/TheShadow29/fairseq.git
cd fairseq
git checkout fseq_v0.9_pyt1.7
pip install -e .

  4. Install coco-caption and cocoapi

    cd $ROOT
git clone https://github.com/cocodataset/cocoapi.git
cd cocoapi/PythonAPI
make

    cd $ROOT
git clone https://github.com/TheShadow29/coco-caption.git

    No additional steps are needed, only need to include coco-caption in path.

After this you should be able to run everything on this repo.