Who said a catwalk can't be a bridge?
Got an R-based model looking to strut its style?
ArcGIS has an R bridge, new scientific python libraries, and multi-dimensional support. This session will work with you to showcase your models and science on the research runway. Working together in advance and then together in the session, we'll look at the models through these different tools, and through the bridge show how to broaden community access to your research and results.
The idea is to connect up some of your data, models, and methods to the out of the box tools to see what new analysis, visualizations, and end-user collaborations can result.
Who Said a Catwalk Can't Be a Bridge?
Let’s talk about the R-Bridge!
How this session came to be?
Thinking about R
Then about models
Then about the R “bridge”
And the theme of sharing and collaborating research
Models and the R Bridge
You can do a lot of work in R, why would you want to connect to ArcGIS in this sort of workflow?
Answer: Both communities can benefit through this sort of collaboration.
Research thrives with different ideas and contributors.
How do you get the feedback and momentum to share, show, and build?
Learn from audience
Understand how to facilitate better science interoperability for the community
Find interested collaborators
Identify where the “sweet spots” for cross platform operations
Roaming mic about who is interested in this talk and why.
Generally: ESRI Employees
LP-DAAC on behalf of users
Broadly to find collaboration
The science that we do is driven largely through Data and Observations.
The domains are diverse, and the GIS can be used to integrate all of these data and domains through sophisticated techniques, tools, and data management practices.
Question for LP DAAC: do you spend a lot of time working with commercial products?
Largely around helping users get data to where they want to start (format conversion).
The computational methods that are useful to this science are varied and diverse as well. GIS can be useful to improve sharing this data, on one single platform.
ArcGIS – a complete web GIS platform deployable on-premises, online, or both (and on desktops as well).
Data -> modeling -> data manipulations -> scientific products -> … -> ArcGIS platform -> end users
ArcGIS has a variety of spatial analysis tools, and is rapidly developing additional toolkits.
Data science in earth science uses a variety of languages and tools.
ArcGIS can work well with many of them, including Python and R.
Question about certain ArcGIS-Python tools (using SciPy). It turns out the software can take care of many of these things on the fly!
Many reasons. Statistical language, widely used, free, etc, etc.
There are performance issues,
Not a general-purpose language
Lacks a purely UI mode of interaction
R – ArcGIS Bridge
The connections between ArcGIS and R can be particularly useful to cross barriers between developers and users. It takes advantage of R AND of ArcGIS.
Why would a user actually want to use the R-ArcGIS Bridge?
R developers can move data easily between R and ArcGIS rapidly, and easily share it with the community who doesn’t necessarily user R.
ArcGIS users can work with tools developed in R WITHOUT needing to work in R.
ArcGIS is a tool with a vast number of spatial tools that are already vetted well, making it very useful to quickly work with these sort of tools, making analysis easier.
R Integration (R-ArcGIS Bridge) examples:
ArcGIS developers can create a custom tool/toolboxes that integrate ArcGIS and R
ArcGIS users can access R code
R users can access Geospatial data managed in traditional GIS ways
The data types in ArcGIS and R are handled differently, likely related to the differing specificity of the tools.
How do you install, access, and work with the R-Bridge from either ArcGIS or R? This is a live demo; please check the recording (slides 32-40).
What is the role of style guiders here, and is there a specific guide that needs to be followed?
Answer: No, not yet.
There is a short plug at the end for the various tools, user communities, and collaborations available for R, ArcGIS, and the bridge.