Skip to content

Latest commit

 

History

History
36 lines (22 loc) · 2.61 KB

README.MD

File metadata and controls

36 lines (22 loc) · 2.61 KB
Raw

Image

Introduction

This repository contains iPython notebook tutorials that explain how to access the iSDAsoil data. For more information about isdasoil, visit https://www.isda-africa.com/isdasoil.

Setup

There are a few different ways to get an iPython notebook up and running. We have created configs for as many options as feasible. You have 2 main options - running a Jupyter Notebook with a cloud provider (e.g. Google or AWS), or running the tutorial locally.

Running on a cloud provider

Option 1: Google colab

Google colab is a free service operated by Google to host iPython notebooks. The only requirement is a Google account. The advantage of this approach is that all requirements are automatically installed, so you can just open the notebook and get started straight away. Note that because this is a cloud service, any data that is downloaded in code will then need to be downloaded locally, should you wish to access the data on your machine.

To open this tutorial in colab, visit: https://colab.research.google.com/github/iSDA-Africa/isdasoil-tutorial/blob/main/iSDAsoil-tutorial.ipynb. Be sure to run the first cell that installs the requirements-googlecolab.txt file.

Option 2: AWS Sagemaker

Equally as easy to set up, the only requirement is an AWS accout. Follow the link below to set up a hosted Jupyter Notebook on AWS Sagemaker, with all requirements preinstalled. Setup can take a few minutes. Once the notebook is set up, be sure to select the isdasoil-tutorial Python kernel.

https://eu-west-2.console.aws.amazon.com/cloudformation/home?region=eu-west-2#/stacks/quickcreate?templateUrl=https%3A%2F%2Fisdasoil.s3-us-west-2.amazonaws.com%2Fsagemaker-config.yml&stackName=isdasoil-tutorial

Running locally

Option 3: conda package manager

  • For those needing to run the iPython locally, the easiest way is to use the environment manager conda. We've created a conda environment.yml config file, with all the requirements specified. Once conda is installed, it's as simple as running the command conda env create -f environment.yml in the directory where the environment.yml file is located.

Option 4: pip + virtualenv

  • The final option is to use pip with a virtual environment. This relies on external dependencies such as the correct version of the GDAL library to be installed. Run the following commands in the folder where you've downloaded the repo to set up the virtual environment:
virtualenv venv
source venv/bin/activate
pip install -r requirements.txt