# AIStream-Peelout/flow-forecast

Deep learning PyTorch library for time series forecasting, classification, and anomaly detection (originally for flood forecasting).

Repository: https://github.com/AIStream-Peelout/flow-forecast
Canonical: https://ross.abutalabs.com/products/flow-forecast
Homepage: https://flow-forecast.atlassian.net/wiki/spaces/FF/overview
Language: Python
License: GPL-3.0
License Family: copyleft
Topics: deep-learning, pytorch, time-series-forecasting, time-series, transfer-learning, deep-neural-networks, transformer, forecasting, lstm, time-series-regression, state-of-the-art-models, anomaly-detection, time-series-analysis, hacktoberfest
Last push: 2026-08-13T06:20:24+00:00

## Health v2 (maintenance only)
Score: 66/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 97, release rhythm 8, longevity 100
- inputs: {"age_days": 2575, "days_push": 20, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2292, forks 300 (observed 2026-08-28T04:06:34.985653+00:00)

## What it is
Flow Forecast is a PyTorch-based deep learning framework for time series forecasting, classification, and anomaly detection. It offers state-of-the-art models (transformers, LSTMs, GRUs, ODEs), interpretability metrics, cloud integration, and model serving, originally built for flood forecasting.

## Use cases
- forecast river flow and floods with deep learning
- train transformer models on multivariate time series
- detect anomalies in time series data
- apply transfer learning to time series regression
- benchmark state-of-the-art time series models
- serve trained forecasting models in production

## When to choose
- you need a PyTorch framework specifically for time series forecasting
- you want transformer-based time series models out of the box
- you need interpretability metrics and model serving in one framework

## When to avoid
- you need classical statistical forecasting like ARIMA or Prophet
- you work outside Python or PyTorch ecosystems
- you need a actively developed project with frequent updates

## Facets
- artifact type: library
- maturity: maintenance
- function: deep-learning, machine-learning, benchmarking
- domain: time-series, machine-learning, deep-learning
- platform: python
- tags: pytorch, time-series-forecasting, transformer, lstm, anomaly-detection, flood-forecasting

## Member repositories
- AIStream-Peelout/flow-forecast (main) score 66

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:34.985653+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 2026-08-30T02:40:39.265585+00:00, confidence not recorded.
  - readme: https://github.com/AIStream-Peelout/flow-forecast (fetched 2026-08-28T04:06:34.985653+00:00, sha 603959e02397)
  - homepage: https://flow-forecast.atlassian.net/wiki/spaces/FF/overview (fetched 2026-08-29T10:20:50.045118+00:00, sha 44136fa355b3)
- Data as of 2026-08-30T08:39:29.467469+00:00.