Skip to content

Latest commit

 

History

History
48 lines (29 loc) · 1.32 KB

INSTALL.md

File metadata and controls

48 lines (29 loc) · 1.32 KB
Raw

Installation Instructions

This document covers how to install pycls and its dependencies.

  • For general information about pycls, please see README.md

Requirements:

  • NVIDIA GPU, Linux, Python3
  • PyTorch, various Python packages; Instructions for installing these dependencies are found below

Notes:

  • pycls does not currently support running on CPU; a GPU system is required
  • pycls has been tested with CUDA 9.2 and cuDNN 7.1

PyTorch

To install PyTorch with CUDA support, follow the installation instructions from the PyTorch website.

pycls

Clone the pycls repository:

# PYCLS=/path/to/clone/pycls
git clone https://github.com/facebookresearch/pycls $PYCLS

Install Python dependencies:

pip install -r $PYCLS/requirements.txt

Set up Python modules:

cd $PYCLS && python setup.py develop --user

Datasets

pycls finds datasets via symlinks from pycls/datasets/data to the actual locations where the dataset images and annotations are stored. For instructions on how to create symlinks for CIFAR and ImageNet, please see DATA.md.

Getting Started

Please see GETTING_STARTED.md for basic instructions on training and evaluation with pycls.