Attribute aggregation/transformation + plotting & evaluation analyses

[glitt13]

An approach to aggregate and transform existing attribute data to create new attribute data
Additionally, create & save plots that visualize results and aid in algorithm performance

Additions

• xssa_attrs_tform.yaml: The example configuration file describing how variables are aggregated and transformed.
• fs_tfrm_attrs.py : this is the main processing script that calls functions inside tfrm_attr.py
• tfrm_attr.py : contains new functions for the fs_proc package
• test_tfrm_attr.py: unit tests
• fs_attrs_miss.R: Reads in missing comid-attributes file sometimes generated during fs_tfrm_attrs.py and attempts to find missing attribute data. In case missing attribute data exist, this Rscript is called inside the fs_tfrm_attrs.py to attempt to acquire data and use it for attribute transformation.
• principal component analysis based on predictor dataset and response variable, including a generation of PCA plot: pca_stdscaled_tfrm, plot_pca_stdscaled_tfrm, plot_pca_stdscaled_cumulative_var, std_pca_plot_path, functions comprehensively summarized in the plot_pca_save_wrap wrapper
• random forest feature importance, all relevant functions comprehensively summarized in the save_feat_imp_fig_wrap wrapper
• algorithm evaluation using the learning_curve analysis, including the AlgoEvalPlotLC class, with functions comprehensively summarized with the plot_learning_curve_save_wrap wrapper
• Create & save predicted vs observed regression plot, comprehensively summarized in the plot_pred_vs_obs_wrap wrapper
• Generate map of predicted response variables, with all relevant functions comprehensively summarized in the plot_map_pred_wrap wrapper
• Generate map of best prediction across multiple datasets, with all relevant functions comprehensively summarized by the plot_best_algo_wrap wrapper 
• AGU 2024 analysis scripts: the ealstm analyses, such as /scripts/analysis/fs_proc_viz_best_ealstm.py and the more-formal /scripts/eval_ingest/ealstm/proc_ealstm_agu24.py plus associated config files in the ealstm/ directory
• Created an updated algo training/testing evaluation script, fs_proc_algo_viz.py as an updated version of fs_proc_algo.py with new evaluation and plotting features

Removals

Changes

• Converted scripts/config/attr_gen_camels.R from hard-coding into a generalizable form that uses the config file scripts/config/attr_gen_camels_config.yaml

Testing

1. Unit testing with test_tfrm_attr.py has been challenging to implement under a normal unittest package approach owing to a mysterious dask.dataframe as dd error. Implemented a work-around that partially tests this package by nixing most instances of using classes.

Screenshots

Notes

Todos

Checklist

• PR has an informative and human-readable title
• Changes are limited to a single goal (no scope creep)
• Code can be automatically merged (no conflicts)
• Code follows project standards (link if applicable)
• Passes all existing automated tests
• Any change in functionality is tested
• New functions are documented (with a description, list of inputs, and expected output)
• Placeholder code is flagged / future todos are captured in comments
• Visually tested in supported browsers and devices (see checklist below 👇)
• Project documentation has been updated (including the "Unreleased" section of the CHANGELOG)
• Reviewers requested with the Reviewers tool ➡️

Testing checklist

Target Environment support

• Windows
• Linux
• Browser

Accessibility

• Keyboard friendly
• Screen reader friendly

Other

• Is useable without CSS
• Is useable without JS
• Flexible from small to large screens
• No linting errors or warnings
• JavaScript tests are passing