Skip to content

Latest commit

 

History

History
59 lines (39 loc) · 1.93 KB

README.md

File metadata and controls

59 lines (39 loc) · 1.93 KB

Use

Parser Script

This script executes a collection of parsers on multiple datasets and saves the results to a pickle file for each dataset and a merged file of all datasets.

Output

The script runs the relevant parser for each dataset specified. The parsed results are then saved in a pickle file in your specified save path. The data for each dataset is in the shape:

household : { appliance : {dataframe with timestamps and values in watts} }

Supported Datasets

Currently, the script supports the following datasets:

  • DEDDIAG
  • DEKN
  • DRED
  • ECO
  • ENERTALK
  • HEART
  • HES
  • HUE
  • IAWE
  • IDEAL
  • LERTA
  • REDD
  • REFIT
  • SustDataED2
  • SustData
  • UCIML
  • UKDALE
  • PRECON
  • EEUD

Load Profile Script

This script processes multiple datasets and calculates the daily, weekly, and monthly load profiles.

Output

The script generates load profiles for each device and aggregate consumption in each household in all the datasets. These are saved in a pickle file in your specified save path. In addition, a merged load profile across all datasets is also generated and saved.

Generate metadata

Output

This script generates a pickle file containing average daily consumption in kWh and an average on/off event consumption in kWh for each household and each device and stores it in the specified save path.

Reset database

This script will delete the database at the specified url in the environment variable DATABASE_URL .env file and recreate it with the provided data. Make sure you have the correct database url in the .env file and the data files in the correct location before running the script.

  • The metadata file should be named residential_metadata.parquet and should be in the data_path folder.
  • The loadprofiles file should be named merged_loadprofiles.pkl and should be in the loadprofiles_path folder.
  • The consumption data file should be named consumption_data.pkl and should be in the consumption_data_path folder.