-
Notifications
You must be signed in to change notification settings - Fork 1.1k
Spatial data science with the Python API an orientation guide
We hear you. Spatial data science is exciting and you understand its potential, but where do you start? Below is a path that you can follow.
Before you learn about analyzing spatial data, you need to get familiar with your analysis environment, the data management, and visualization environment. This is your Geospatial cloud that we are talking about. Below are some resources gathered from http://learn.arcgis.com site. These will act as good foundations to build your knowledge about spatial analysis
- Getting familiar ArcGIS Online - https://learn.arcgis.com/en/paths/getting-started-path-for-teachers/
- part 2: getting to know arcgis online - https://learn.arcgis.com/en/paths/try-arcgis-online/
- mapping and visualization - https://learn.arcgis.com/en/paths/mapping-and-visualization/
- data analysis using arcgis online - https://learn.arcgis.com/en/paths/data-analysis/
- sharing your analysis - https://learn.arcgis.com/en/paths/sharing-and-collaboration/
The lessons so far don’t involve any coding. But, the ones below do. These are bite sized tutorials of accomplishing some basic tasks using the Python API - called Python Dev Labs. It would be good to try them out, particularly the challenge sections.
Below are some examples that demonstrate how you perform analysis not just using Python API, but using some of the popular data analysis and plotting packages in the scientific Python ecosystem:
- Fighting California forest fires using spatial analysis
- Site suitability - Finding suitable spots for placing heart defibrillator equipments in public
- Analyzing raster data - Identifying natural and accessible areas using weighted overlay analysis
- Network analysis - Constructing drive time based service areas
- Finding nearest facilities - Find hospitals closest to an incident
Find more of these examples at: Python API - sample notebooks. To learn the concepts behind these notebooks, visit Python API - guide
Now you are familiar with the Geospatial Cloud, the ArcGIS API for Python, Jupyter Notebooks and a host of data analysis and plotting libraries in the Python ecosystem. Now it is time to dig deeper into the core concepts of spatial analysis. Below are some great resources to read:
Esri Guide to Spatial Analysis
- Volume 1 - Geographic Patterns and Relationships,
- Volume 2 - Spatial Measurements and Statistics,
- Volume 3 - Modeling, Suitability, Movement and Interaction
Below are the books you need to read to become familiar with the core statistical concepts behind many of the machine learning algorithms used:
- Introduction to Statistical Learning
- Follow that up with advanced Elements of statistical learning
While reading these books, it is good to follow along by coding the examples in Python. Checkout ISLR-Python GitHub repo for some inspiration.