# oliverguhr/transformer-time-series-prediction

proof of concept for a transformer-based time series prediction model

Repository: https://github.com/oliverguhr/transformer-time-series-prediction
Canonical: https://ross.abutalabs.com/products/transformer-time-series-prediction
Language: Python
License: MIT
License Family: permissive
Last push: 2023-02-20T14:48:50+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2379, "days_push": 1290, "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 1453, forks 259 (observed 2026-08-28T04:04:46.548693+00:00)

## What it is
A proof-of-concept repository with two PyTorch transformer models for single-step and multi-step time series prediction. It is educational demo code, with the author directing serious use to the flow-forecast package.

## Use cases
- learn how to build a transformer for time series forecasting
- predict future values from historical time series data
- example of single-step and multi-step prediction in PyTorch
- starting point for transformer-based forecasting experiments
- understand how transformers capture long-term trends in sequences

## When to choose
- you want readable reference code for transformer time series models
- you are learning PyTorch and sequence prediction
- you need a minimal demo of multi-step forecasting

## When to avoid
- you need production-ready, efficient, or bug-free forecasting
- you want a maintained library with full features - use flow-forecast instead
- you need guaranteed correctness for critical workloads

## Facets
- artifact type: learning-resource
- maturity: maintenance
- function: machine-learning, deep-learning
- domain: time-series, deep-learning, machine-learning
- platform: python
- tags: transformer, pytorch, time-series-forecasting, proof-of-concept

## Member repositories
- oliverguhr/transformer-time-series-prediction (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:46.548693+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 2026-08-30T04:35:45.499254+00:00, confidence not recorded.
  - readme: https://github.com/oliverguhr/transformer-time-series-prediction (fetched 2026-08-28T04:04:46.548693+00:00, sha 1c7cfd204d66)
- Data as of 2026-08-30T08:39:29.467469+00:00.