shibing624/dialogbot

dialogbot, provide search-based dialogue, task-based dialogue and generative dialogue model. 对话机器人,基于问答型对话、任务型对话、聊天型对话等模型实现,支持网络检索问答,领域知识问答,任务引导问答,闲聊问答,开箱即用。

61
/ 100
Established

Implements three dialogue paradigms—retrieval-based (using TFIDF/BM25 similarity matching against local QA corpora or web search results from Baidu/Bing), task-oriented (via End-to-End Memory Networks on BABI datasets), and generative (GPT-2 or seq2seq models trained on conversational corpora)—allowing developers to select or combine approaches based on use case. Built on PyTorch and Hugging Face Transformers with pre-trained Chinese dialogue models available, supporting multi-turn conversation tracking and configurable decoding parameters (top-k, top-p, repetition penalties) for response generation tuning.

330 stars and 39 monthly downloads. No commits in the last 6 months. Available on PyPI.

Stale 6m No Dependents
Maintenance 0 / 25
Adoption 14 / 25
Maturity 25 / 25
Community 22 / 25

How are scores calculated?

Stars

330

Forks

63

Language

Python

License

Apache-2.0

Last pushed

Apr 23, 2024

Monthly downloads

39

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/shibing624/dialogbot"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.