rasac 2.1.1

A GUI for NLU pipeline configurations and model training for Rasa Conversational AIs.

RASA Codeless provides a GUI for NLU pipeline configurations and model training and evaluation for Rasa Conversational AIs.

Features

• Allows configuring, training, and evaluating RASA models efficiently through a React based UI
• Fully supports rasa 2.8.x projects
• Provides an easy-to-use CLI
• Provides an efficient server-based GUI
• Requires minimum machine learning expertise and RASA framework specific knowledge
• Learning curve insights for non-experts

What's Cooking?

• Training data management
• Story designer
• Custom config file upload
• NLU file upload
• CDD (Interactive Learning)
• Rule designer
• Conversation flow designer
• Custom NLU pipeline components attaching
• Cross-validation and F1-score based performance evaluation
• Automated validation dataset size configuration
• User-friendly model and deployment management
• One-click deploy
• Docker support for model deployment
• Custom MongoDB botstore support
• Omni-channel support

Limitations and Known Issues

• Does not support early stopping
• Training ETA not available
• Evaluations are limited to DIET Classifier for now

📒 Docs: https://rasa-codeless.github.io
📦 PyPi: https://pypi.org/project/rasac/2.1.1/
🪵 Full Changelog: here