sfu-bigdata / range-driver-tutorials Goto Github PK

Write Loading & Processing Tutorial

Write the tutorial to go through the process of loading & processing detection data. I think we can start with loading through the instantiation of a Detections object, but then explain what processing is done by default (calculate detection rate, event vs binning view, groups, environmental data).

At the end of this notebook, we may want to include how to write the data out for (a) analyzing in another script or (b) to ensure you don't have to re-run the detections processing.

Write the Data Visualization Tutorial

Go through the built-in reporting & visualization functionality (boundary map, bivariate distribution plots, heatmaps) and briefly mention other tools that could be used for other visualizations (seaborn, etc).

I think this should go through the defaults of the functions, but then also explain slight variations. For example, how do we change which variables are being compared in the bivariate distribution plots?

Write Configuration Tutorial

This tutorial should focus on how to read, write, and manage the configuration files that are critical to this package. Some key points:

• We should mention (but not dwell on) that our config files are written in YAML
• This should include a table or well-structured bullets that detail each of the possible fields in the configuration files
• Talk about how to use multiple configuration files and why you might want to
• When we describe the boundaries field, we should also explain how a user could go about determining their source. Note: this will need to be updated with Steven's work on multiple boundaries for different data sources.

Write Quickstart tutorial.

This should be a very minimal example that runs through the entire workflow (read in config, create & process detections, view detections, generate basic report, etc). Some specific points:

• We can likely skip the mapping of nodes, since it is primarily used for determining appropriate boundary settings. However, we could still use map the receiver locations. , since it is primarily used . . We should use the abstracted/brief methodology for this.
• Include a line to show how the combined dataframe could be exported for use in other notebooks like Python and R.
• Expectation at this point is that readers will have already installed range_driver successfully.
• Focus on brevity

Write Data Setup Tutorial

Unsure what to call this one at this point... preparation? files? setup?

Regardless, we need a tutorial that goes through user-provided data files, explaining what information is required/expected in each file type. Some other notes:

• We should establish the folder and file structure for detection and environmental data sources (the folder structure is less important I think... since file locations will all be specified in the config).
• Go through each user-provided file --> Detection Events, Metadata, & Custom Environmental Data

Convert the current README table of contents into a series of notebook skeletons. To start, we don't need to fill in the content of these notebooks (that will be the next step) but just create the notebooks and their structure with headings.