project workflow

< Back

The jobs are meant to run autonomously, but for demo purposes we can manually run the ETL and feature selection jobs from the notebooks/nfl_load_nflverse_data_demo.ipynb notebook.

• nfl_load_nflverse_data_demo.ipynb - demos manually running the load and build jobs (nfl_00 - 01)

• nfl_perform_feature_selection_demo.ipynb - demos manually running weekly stats and feature selection jobs (nfl_02 - 04)

Job	Stage	Description
nfl_main.py	orchestration	runs the ETL and feature selection for the win/loss model
nfl_00_load_nflverse_data.py	ETL	downloads the nflverse data into local storage
nfl_01_build_nfl_database.py	ETL	builds the nfl database from the nflverse data
nfl_02_prepare_weekly_stats.py	Experiment 2	merges metrics from all datasets into a single stats dataset
nfl_03_perform_feature_selection.py	Experiment 2	performs feature selection on the nfl data
nfl_04_merge_game_features.py	Experiment 2	merge our features with the core nfl play actions dataset

Finally, we can run the Experiment 2 notebook