Initial lib was developed during Open AI Challenge with Omdena. Learn more about the challenge at here

This is a fork from the original (and on-going) HOT project, created with the aim to analyse its performance and to work on improving it.

The workflow and notes behind this project is in this HackMD (ask for access if needed).

Note:

• write about the env file
• write about the metric generation

hot_fair_utilities is a collection of utilities which contains core logic for model data preparation, training and postprocessing. It can support multiple models, currently ramp is supported.

1. To get started clone this repo first :

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

2. Setup your virtualenv with python 3.8 ( Ramp is tested with python 3.8 )

3. Install tensorflow 2.9.2 from [here] (https://www.tensorflow.org/install/pip) According to your os

4. Copy your basemodel : Basemodel is derived from ramp basemodel

   git clone https://github.com/radiantearth/model_ramp_baseline.git
   

5. Clone ramp working dir

   git clone https://github.com/kshitijrajsharma/ramp-code-fAIr.git ramp-code
   

6. Copy base model to ramp-code

   cp -r model_ramp_baseline/data/input/checkpoint.tf ramp-code/ramp/checkpoint.tf
   

7. Install native bindings

   • Install Numpy

     pip install numpy==1.23.5
     

   • Install gdal .

     for eg : on Ubuntu

     sudo add-apt-repository ppa:ubuntugis/ppa && sudo apt-get update
     sudo apt-get install gdal-bin
     sudo apt-get install libgdal-dev
     export CPLUS_INCLUDE_PATH=/usr/include/gdal
     export C_INCLUDE_PATH=/usr/include/gdal
     pip install --global-option=build_ext --global-option="-I/usr/include/gdal" GDAL==`gdal-config --version`        
     

     on conda :

     conda install -c conda-forge gdal
     

   • Install rasterio

     for eg: on ubuntu :

     sudo apt-get install -y python3-rasterio
     

     on conda :

     conda install -c conda-forge rasterio
     

8. Install ramp requirements

   Install necessary requirements for ramp and hot_fair_utilites

   cd ramp-code && cd colab && make install  && cd ../.. && pip install -e .
   

Create from env fle

conda env create -f environment.yml

Create your own

conda create -n fAIr python=3.8
conda activate fAIr
conda install -c conda-forge gdal
conda install -c conda-forge geopandas
pip install pyogrio rasterio tensorflow
pip install -e hot_fair_utilities

You can run package_test.ipynb to your pc to test the installation and workflow with sample data provided