dennybritz/chatbot-retrieval resource

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1570 stars · 665 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A reference implementation of the Dual LSTM Encoder model for retrieval-based dialog response selection, built in TensorFlow and trained on the Ubuntu Dialogue Corpus. It accompanies a WildML blog post on deep learning for chatbots and includes scripts for training, evaluation, and prediction.

Use cases

• learn how to build a retrieval-based chatbot with deep learning
• implement a dual LSTM encoder for dialog response selection
• train a response ranking model on the Ubuntu Dialog Corpus
• understand how to evaluate a next-utterance classification model
• study a worked TensorFlow example for conversational AI

When to choose

• you want to learn retrieval-based dialog models from a well-known tutorial implementation
• you need a baseline dual encoder architecture for response selection research
• you are working through the Ubuntu Dialogue Corpus paper experiments

When to avoid

• you need a production-ready chatbot framework or maintained library
• you require modern TensorFlow 2.x or PyTorch compatibility, since the code targets TensorFlow 0.9-era APIs
• you want a generative conversational model rather than retrieval-based response ranking

Facets

learning-resource · maturity abandoned

machine-learning deep-learning rag chatbot deep-learning chatbots machine-learning tutorials python cross-platform dual-lstm-encoder tensorflow ubuntu-dialog-corpus retrieval-based-chatbot response-selection reference-implementation jupyter-notebook natural-language-processing

Member repositories

For agents

markdown · JSON · MCP: product_card(name="dennybritz/chatbot-retrieval")