-
ARLR_exog.py: All functions related to ARLR with exogenous variables forecast
-
data_prep.py: Contains the data preparation (reading CSVs and preparing basic dataframes), some code specific data handling is performed by ARLR_regressors function within ARLR_exog.py
-
forecast_ARLR_exog.py: Wrapper function that parsing of command line arguments. ARLR_module_exog.py within forecast_ARLR_exog.py call the ARLR necessary functions.
-
out_format.py: Contains functions for output file preparation for accuweather or CDC.
Usage: python scripts/forecast_ARLR_exog.py -b 201840 -w 4 --out_folder dump/ --ground_truth data/state/ILINet.csv -v --CDC 0 --st_fips data/state_fips.csv --out_state output/predictions.txt --accu_data data/data-aw-cumulative_20191018_1620-weekly-state.csv
Following are the various options for argument: For Accuweather, --out_state and --st_fips are mandatory. If exogenous regressors are required, --accu_data must be included with path to the weather data csv file. usage: forecast_ARLR_exog.py [-h] [-b FORECAST_FROM] -w WEEKS [--out_state OUT_STATE] [--out_county OUT_COUNTY] [--ground_truth GROUND_TRUTH] [--region_type REGION_TYPE] [--st_fips ST_FIPS] [--county_ids COUNTY_IDS] [--end_date END_DATE] --CDC CDC [--test TEST] [-v] [-l LOG] -o OUT_FOLDER [--accu_data ACCU_DATA] [--ght_data GHT_DATA]