# google/seq2seq

A general-purpose encoder-decoder framework for Tensorflow

Repository: https://github.com/google/seq2seq
Canonical: https://ross.abutalabs.com/products/seq2seq
Homepage: https://google.github.io/seq2seq/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: tensorflow, translation, machine-translation, neural-network, deeplearning
Archived: true
Last push: 2020-10-15T08:00:39+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3471, "days_push": 2148, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 5618, forks 1283 (observed 2026-08-28T04:09:23.068641+00:00)

## What it is
A general-purpose encoder-decoder (seq2seq) framework built on TensorFlow, supporting tasks like machine translation, text summarization, conversational modeling, and image captioning. Models and training pipelines are configured via YAML for reproducibility, with modular, extensible architecture code.

## Use cases
- train a neural machine translation model
- build a text summarization model
- create a conversational chatbot model
- train an image captioning model
- experiment with attention mechanisms and encoder architectures
- reproduce NMT research experiments from YAML configs

## When to choose
- you need a configurable encoder-decoder framework on TensorFlow 1.x
- you want to reproduce the 'Massive Exploration of Neural Machine Translation Architectures' paper
- your seq2seq task fits encode-input/decode-output framing and you want modular extensibility

## When to avoid
- you use TensorFlow 2.x or PyTorch - the project is unmaintained and tied to old TensorFlow
- you need state-of-the-art transformer or LLM-based models
- you need maximum training performance or production-grade support

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, deep-learning, nlp
- domain: machine-learning, deep-learning
- platform: python
- tags: tensorflow, encoder-decoder, seq2seq, machine-translation, text-summarization, image-captioning, natural-language-processing, gpu

## Member repositories
- google/seq2seq (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:23.068641+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 2026-08-29T17:55:24.010878+00:00, confidence not recorded.
  - readme: https://github.com/google/seq2seq (fetched 2026-08-28T04:09:23.068641+00:00, sha 1b2e09075bb5)
  - homepage: https://google.github.io/seq2seq/ (fetched 2026-08-29T08:50:05.004908+00:00, sha 746d89200f3b)
- Data as of 2026-08-30T08:39:29.467469+00:00.