# 920232796/bert_seq2seq

pytorch实现 Bert 做seq2seq任务，使用unilm方案,现在也可以做自动摘要，文本分类，情感分析，NER，词性标注等任务,支持t5模型，支持GPT2进行文章续写。

Repository: https://github.com/920232796/bert_seq2seq
Canonical: https://ross.abutalabs.com/products/bert_seq2seq
Language: Python
License: Apache-2.0
License Family: permissive
Topics: bert, seq2seq, ner, crf, text-classification, unilm, pytorch, roberta, autotitle, gpt2, t5-model
Last push: 2022-06-18T12:34:11+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": 2364, "days_push": 1537, "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 1308, forks 207 (observed 2026-08-28T04:04:19.321856+00:00)

## What it is
A lightweight PyTorch framework for fine-tuning pretrained language models (BERT, RoBERTa, Nezha, GPT2, T5, BART) on Chinese NLP tasks using the UniLM seq2seq approach. It supports sequence-to-sequence generation, text classification, sequence labeling with optional CRF, relation extraction, and similar-sentence generation.

## Use cases
- generate automatic summaries or titles from Chinese text
- fine-tune BERT for named entity recognition with CRF
- train a seq2seq model to write poems or couplets
- do sentiment analysis or text classification with BERT
- continue writing articles with GPT2
- extract relation triples from text
- generate similar sentences with SimBERT

## When to choose
- you need a simple, lightweight PyTorch setup for Chinese NLP fine-tuning
- you want seq2seq tasks like summarization or title generation with BERT-style models
- you want quick examples for NER, classification, and relation extraction in one framework

## When to avoid
- you need production-grade distributed training or the latest model architectures
- you primarily work with English-only NLP pipelines
- you prefer using Hugging Face Transformers directly with its broader ecosystem

## Facets
- artifact type: framework
- maturity: maintenance
- function: machine-learning, nlp, llm-training, deep-learning
- domain: machine-learning, deep-learning
- platform: python
- tags: bert, seq2seq, unilm, t5, gpt2, ner, text-classification, pytorch, chinese-nlp, text-summarization, natural-language-processing

## Member repositories
- 920232796/bert_seq2seq (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:19.321856+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 2026-08-30T04:50:02.419645+00:00, confidence not recorded.
  - readme: https://github.com/920232796/bert_seq2seq (fetched 2026-08-28T04:04:19.321856+00:00, sha 0d9b6fa4cd67)
  - registry_pypi: https://pypi.org/pypi/bert_seq2seq/json (fetched 2026-08-29T12:08:18.087946+00:00, sha dfd399a0a2a9)
- Data as of 2026-08-30T08:39:29.467469+00:00.