hot-fair-utilities 2.0.12

Utilities for AI - Assisted Mapping fAIr

hot_fair_utilities ( Utilities for AI Assisted Mapping fAIr )

Initially lib was developed during Open AI Challenge with Omdeena. We frequently do AI challenges with community !

Prerequisties

• Install gdal-python and numpy array

hot_fair_utilities Installation

Installing all libraries could be pain so we suggest you to use docker , If you like to do it bare , You can follow .github/build.yml

Clone repo

git clone https://github.com/hotosm/fAIr-utilities.git

Navigate to fAIr-utilities:

cd fAIr-utilities

Build Docker

docker build --tag fairutils .

Run Container with default Jupyter Notebook , Or add bash at end to see terminal

docker run -it --rm --gpus=all  -p 8888:8888 fairutils

[Optional] If you have downloaded RAMP already , By Default tf is set as Ramp_Home , You can change that by attaching your ramp-home volume to container as tf

if not you can skip this step , Ramp code will be downloaded on package_test.ipynb

-v /home/hotosm/fAIr-utilities:/tf

Test inside Docker Container

docker run -it --rm --gpus=all  -p 8888:8888 fairutils bash

python test_app.py

Test Installation and workflow

You can run package_test.ipynb on your notebook from docker to test the installation and workflow with sample data provided , Or open with collab and connect your runtime locally

Get started with development

Now you can play with your data , use your own data , use different models for testing and also Help me Improve me !

Version Control

Follow Version Control Docs to publish and maintain new version

master --- > Dev
Releases ---- > Production