Skip to content

INSTALL.md

Latest commit

 

History

History
86 lines (66 loc) · 2.29 KB

INSTALL.md

File metadata and controls

86 lines (66 loc) · 2.29 KB

Installation

Requirements

  • Linux
  • Python 3.5+
  • PyTorch 1.1
  • CUDA 9.0+
  • NCCL 2+
  • GCC 4.9+
  • mmcv

We have tested the following versions of OS and softwares:

  • OS: Ubuntu 16.04/18.04 and CentOS 7.2
  • CUDA: 9.0/9.2/10.0
  • NCCL: 2.1.15/2.2.13/2.3.7/2.4.2
  • GCC: 4.9/5.3/5.4/7.3

Install Aerialdetection

a. Create a conda virtual environment and activate it. Then install Cython.

conda create -n fod python=3.7 -y
source activate fod

conda install cython

b. Install PyTorch stable or nightly and torchvision following the official instructions. An example is given below:

conda install pytorch==1.5.0 torchvision==0.6.0 cudatoolkit=10.2 -c pytorch

c. Clone this repository (Skip this step if it exists locally).

git clone https://github.com/NCL-PYRAMID/PYRAMID-object-detection.git
cd PYRAMID-object-detection

d. Compile cuda extensions.

./compile.sh

e. Install AerialDetection (other dependencies will be installed automatically).

pip install -r requirements.txt
python setup.py develop
# or "pip install -e ."

Note:

  1. It is recommended that you run the step e each time you pull some updates from github. If there are some updates of the C/CUDA codes, you also need to run step d. The git commit id will be written to the version number with step e, e.g. 0.6.0+2e7045c. The version will also be saved in trained models.

  2. Following the above instructions, AerialDetection is installed on dev mode, any modifications to the code will take effect without installing it again.