# KEV0143/Adaptive-forecasting-of-electricity-consumption-24-168-720-h-with-load-regime-conditioning

Repository: https://github.com/KEV0143/Adaptive-forecasting-of-electricity-consumption-24-168-720-h-with-load-regime-conditioning
Canonical: https://ross.abutalabs.com/products/adaptive-forecasting-of-electricity-consumption-24-168-720-h-with-load-regime-conditioning
License: Apache-2.0
License Family: permissive
Last push: 2026-07-20T16:37:12+00:00

## Health v2 (maintenance only)
Score: 59/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 93, release rhythm 35, longevity 24
- inputs: {"age_days": 341, "days_push": 44, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1545, forks 0 (observed 2026-08-28T04:05:01.245720+00:00)

## What it is
An open-source tool for adaptive forecasting of hourly electricity consumption over 24, 168, and 720 hour horizons, using load regime conditioning to improve prediction accuracy. It is implemented as a machine learning forecasting library for energy time series.

## Use cases
- forecast hourly electricity consumption for the next day
- predict weekly (168-hour) power demand
- generate monthly (720-hour) load forecasts
- model electricity consumption under different load regimes
- build adaptive energy demand forecasting pipelines
- evaluate multi-horizon load forecasting models

## When to choose
- you need multi-horizon (24/168/720h) electricity load forecasts
- your consumption patterns vary by load regime and you want conditioning on them
- you want an Apache-2.0 licensed forecasting solution for energy time series

## When to avoid
- you need forecasting for domains other than electricity consumption
- you require real-time sub-hourly forecasting granularity
- you need a turnkey commercial forecasting service with support

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, data-science
- domain: machine-learning, data-science, energy
- platform: python, cross-platform
- tags: time-series-forecasting, electricity-load-forecasting, energy, hourly-consumption, regime-conditioning, 24-168-720-hour-horizons, time-series

## Member repositories
- KEV0143/Adaptive-forecasting-of-electricity-consumption-24-168-720-h-with-load-regime-conditioning (main) score 59

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:01.245720+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:30:34.859129+00:00, confidence not recorded.
  - readme: https://github.com/KEV0143/Adaptive-forecasting-of-electricity-consumption-24-168-720-h-with-load-regime-conditioning (fetched 2026-08-28T04:05:01.245720+00:00, sha 4a10fe72a0de)
- Data as of 2026-08-30T08:39:29.467469+00:00.