# philipperemy/keras-tcn

Keras Temporal Convolutional Network. Supports Python and R.

Repository: https://github.com/philipperemy/keras-tcn
Canonical: https://ross.abutalabs.com/products/keras-tcn
Language: Python
License: MIT
License Family: permissive
Topics: keras, tcn, recurrent-neural-networks, deep-learning, machine-learning
Last push: 2026-06-16T12:28:04+00:00

## Health v2 (maintenance only)
Score: 62/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 87, release rhythm 8, longevity 100
- inputs: {"age_days": 3086, "days_push": 78, "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 2012, forks 462 (observed 2026-08-28T04:06:05.188622+00:00)

## What it is
A Keras/TensorFlow implementation of Temporal Convolutional Networks (TCN) with dilated causal convolutions, usable as a drop-in layer alternative to LSTM/GRU. It supports Python and R and is actively tested against recent TensorFlow releases.

## Use cases
- build temporal convolutional networks in keras
- replace lstm with tcn for long time series
- sequence classification with dilated causal convolutions
- time series forecasting model in tensorflow
- train tcn on gpu with tensorflow-metal on macos
- use tcn layer in r with keras

## When to choose
- you need long-memory sequence modeling with stable gradients and parallel training
- you want a Keras layer that swaps in for LSTM/GRU
- you work in Python or R with TensorFlow 2.x

## When to avoid
- you need PyTorch or JAX instead of TensorFlow/Keras
- your sequences are short and an LSTM suffices
- you need pretrained models rather than a layer implementation

## Facets
- artifact type: library
- maturity: active
- function: deep-learning, machine-learning
- domain: deep-learning, machine-learning, time-series
- platform: python
- tags: keras, tcn, temporal-convolutional-network, tensorflow, sequence-modeling, r-language, gpu

## Member repositories
- philipperemy/keras-tcn (main) score 62

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:05.188622+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 2026-08-30T03:01:02.915741+00:00, confidence not recorded.
  - readme: https://github.com/philipperemy/keras-tcn (fetched 2026-08-28T04:06:05.188622+00:00, sha 00e7edb61634)
- Data as of 2026-08-30T08:39:29.467469+00:00.