Skip to content

Latest commit

 

History

History
35 lines (23 loc) · 1001 Bytes

README.md

File metadata and controls

35 lines (23 loc) · 1001 Bytes
Raw

WildSight

Accelerating conservation with AI

header

This codebase was created for a senior design project in COE 374. This project is currently in development.

Setup

There are two main projects within this monorepo. One for training AI models and another for using the models in the browser. The browser application is housed in wild_sight/web.

It's recommended for this code to be run on Ubuntu. As of February 2021, Ubuntu 18.04 is the main development OS, but 20.xx likely works, too. You may also use WSL.

To get the environment setup, run:

python3 -m pip install -r requirements.txt

Training

PYTHONPATH=. wild_sight/train/detection/train.py \
    --config wild_sight/train/configs/vovnet-det.yaml

Web

Change directory into wild_sight/web and find the README in that folder. There you will see the instructions for setting up the web project.