Download
ShoppingAgent is an open source Chinese conversational AI system that allows users to build and train their own chatbot using custom datasets. It provides multiple implementations of chatbot architectures, including traditional Seq2Seq models as well as newer GPT-style approaches, reflecting the evolution of conversational AI techniques. ShoppingAgent is structured to support experimentation across different deep learning frameworks such as TensorFlow, PyTorch, and MindSpore, giving developers flexibility in how they train and deploy models. In addition to core chatbot functionality, the project introduces agent-based capabilities, enabling practical use cases like automated workflows and task-oriented assistants. It also includes support for small language models and local training scripts, making it accessible for users with limited computational resources. ShoppingAgent can be applied to scenarios such as customer service, question answering, and casual conversation.
Features
- Train custom chatbot models using user-provided Chinese datasets
- Supports multiple architectures including Seq2Seq and GPT-style models
- Compatible with TensorFlow, PyTorch, and MindSpore frameworks
- Includes agent-based tools for automation and real-world task execution
- Enables local training with small language models and scripts
- Provides web interface and conversational UI for interaction
Project Samples
