Discover gists

How I do machine learning with geospatial data

I have a couple of AI/ML projects related to mapping things, often conservation related, with remote sensing data. Some details and packages will vary, but the process below describes how I generally approach these types of problems. Some of these tools I have only touched briefly, but I like them, and this is more an outline of how I would like to approach a new project than a retrospective look at my previous work.

We use AWS, so it makes sense to use datasets and services that are already hosted on AWS. The data discovery and loading part of this process would look somewhat different if we were using Azure and Planetary Computer, and very different if we were using GCP and Earth Engine.

Compute

All of my analysis will be done using python on an AWS VM in the same region as my data on S3, Probably using VSCode on Sagemaker or [JupyterLab](https://docs.aws.amazo

	var webpack = require('webpack')
	var ExtractTextPlugin = require("extract-text-webpack-plugin")
	var path = require('path');

	module.exports = {

	entry: "./src/index.js",
	output: {
	path: path.resolve(__dirname, ''), // string
	filename: "[name].js", // string

	// Future versions of Hyper may add additional config options,
	// which will not automatically be merged into this file.
	// See https://hyper.is#cfg for all currently supported options.

	module.exports = {
	config: {
	// default font size in pixels for all tabs
	fontSize: 12,

	// font family with optional fallbacks

	// usage : truthTable((a,b)=>a&&b, 2)
	// 0,0 - 0
	// 0,1 - 0
	// 1,0 - 0
	// 1,1 - 1

	function truthTable(question, argcount) {
	var combinations = binaryCombos(argcount)
	for (var combination in combinations) {
	combination = combinations[combination].reverse()

	#if( $url.query )
	## Only execute when there is a query called action=json. This is our API route more or less.
	## For example: localhost:8090/?action=json.getBlogPosts&pages=Test1\|\|\|Test2
	## This will execute the getBlogPosts Macro with "pages" as parameter
	#if( $url.query.indexOf('action=json.') != -1 )
	#set( $actionIdentifier = 'action=json.' )
	#if( $url.query.indexOf('&') != -1 )
	#set( $action = $url.query.substring($actionIdentifier.length(),$url.query.indexOf('&')))
	#set( $i = "$actionIdentifier$action" )
	#set( $l = $i.length() + 1 )

	<script>
	var versions = [
	#foreach ($v in $versions.available)
	"$v.name",
	#end
	]

	var x = versions.reverse().reduce(function(acc, item, index, items){
	if (acc) {
	return acc

	// location.search = '?a=b&c=d&limit=20'
	var get = new Map(location.search.substr(1).split('&').map((pair)=>{return pair.split('=')}))
	// get = {a:'b', c:'d', limit: 20}

	function rot(text, amount=13) {
	return text.split('').map(letter=>{return String.fromCharCode(((letter.charCodeAt(0)-97+amount) % 25) + 97)}).join('')
	}

	## place somewhere at the bottom before </body> - doesn't matter
	<script>
	window.spaceKey = '$repository.key'
	</script>

	const sqlFormatter = require('sql-formatter')
	const promisify = require('promisify-node')
	const fs = promisify('fs')
	const path = require('path')

	const INDENT = '\t'
	const SCHEMAFILE = path.resolve(__dirname, 'schema.sql')

	(async () => {
	try {