Geospatial Build

Aakash Ahamed
Stanford University
September, 2018

This readme contains instructions to build a geospatial data processing environment which includes:

• Python 3.6
• Earth Engine
• GDAL, Geopandas, Rasterio, Fiona, other geospatial libs
• Numpy, scipy, pandas, other useful python libs
• Tensorflow, Keras, sklearn, other ML libs

This system has been run and tested on MacOS High Sierra 10.13.6

Download Anaconda3

Link to MacOS Anaconda Installers

Clone this repository

git clone https://github.com/kashingtonDC/geospatial_build

Create a new conda virtual environment from the env.yml file

conda env create -f env.yml

This will take a while to install the packages. The name of the environment is 'gis', which is specified in the first line of the env.yml file

Activate the environment:

source activate gis

Note that for newer versions of conda, this command may be conda activate gis. Then (gis) will be prepended to your path.

Now install Tensowflow backend:

pip install --ignore-installed --upgrade https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-1.10.1-py3-none-any.whl

And Keras:

pip install keras

Test everything:

python
import ee
import gdal
import tensorflow as tf
import keras
ee.Initialize() # Supply your EE credentials

That's it!