Webinar: How to Use Integrated Version Control in Rasa X
0 likes257 views
The document discusses integrated version control in Rasa X, which allows teams to version training data and push changes made in Rasa X to a target branch connected to a remote git repository. It provides an overview of getting started with setting up version control by deploying Rasa X, adding an SSH key to GitHub, and connecting Rasa X. The document also discusses how integrated version control connects Rasa X to downstream workflows like automated testing and deployments and brings it in line with software engineering best practices.
Webinar: How to Use Integrated Version Control in Rasa X
- 1. How to Use Integrated Version Control in Rasa X Ty Dunn Product Manager Karen White Developer Marketing Manager
- 2. Agenda HOW TO USE INTEGRATED VERSION CONTROL IN RASA X Intro Rasa X: a toolset for building mission-critical AI assistants ● What’s new: Integrated Version Control Software Development Workflows for Building AI Assistants ● Version control ● Testing ● CI/CD Demo: Integrated Version Control Q&A
- 3. OUR MISSION Empower all makers to create AI assistants that work for everyone
- 4. Rasa X: downloaded in 135 countries
- 5. Rasa X is For Teams Building Mission-Critical Assistants
- 6. Going beyond simple FAQ chatbots Level 1: Notifications Level 2: “Chatbots” - FAQs Level 3: Contextual Assistants Level 4: Personalized Assistants Level 5: Autonomous Organization
- 7. Build minimum viable assistant Improve by talking to the assistant Improve using conversations with real users Improve using conversations with test users Quality of assistant Rasa Open Source (Local) Rasa X (Server) Rasa Open Source is an open source framework for natural language understanding, dialogue management, and integrations. Rasa X is a toolset used to improve a contextual assistant built using Rasa Open Source. Deploy your minimum viable assistant on a server and improve it using Rasa X
- 8. Continually improve your assistant using Rasa X Review conversations and improve your assistant based on what you learn Collect conversations between users and your assistant Ensure your new assistant passes tests using continuous integration (CI) and redeploy it to users using continuous deployment (CD)
- 9. Rasa X was designed for one purpose: To help product teams go from simple bots to advanced AI assistants HOW TO USE INTEGRATED VERSION CONTROL IN RASA X Conversational datasets, specific to the assistant’s task Powerful tools for reviewing past conversations Integration with development workflows Teams building AI assistants need 3 things: With Integrated Version Control, Rasa X now does all 3. 1 2 3
- 10. NEW: Integrated Version Control HOW TO USE INTEGRATED VERSION CONTROL IN RASA X What does it do? ● Creates a two-way sync with a remote git repository ● Allows you to version training data and push changes made in Rasa X to a target branch By extension… ● Connects Rasa X to downstream workflows like automated testing and deployments ● Brings Rasa X in line with software engineering best practices Enable in version 0.23.0+ as an experimental feature
- 11. Overview: Getting Started INTEGRATED VERSION CONTROL 1 Deploy Rasa X Install Rasa X on a server Generate SSH key Create a new SSH key pair on your server 2 Add key to GitHub Add your public key to your repository settings 3 Connect with Rasa X Establish the connection by posting your private SSH key and repository URL to a Rasa X API endpoint 4
- 13. Applications using machine learning still need engineering best practices. HOW TO USE INTEGRATED VERSION CONTROL IN RASA X
- 14. End-to-end workflow for managing assistant updates INTEGRATED VERSION CONTROL
- 15. After you build, run model tests using Rasa CLI HOW TO USE INTEGRATED VERSION CONTROL IN RASA X Split data into a test set, or estimate how well your model generalizes using cross-validation. NLU Model Evaluation $ rasa test nlu -u data/nlu.md --config config.yml --cross-validation Evaluate your trained model on a set of test stories and generate a confusion matrix. Core Evaluation $ rasa test core --stories test_stories.md --out results End to End Evaluation $ rasa test --stories e2e_stories.md --e2e Run through test conversations to make sure that both NLU and Core make correct predictions. See an example using GitHub Actions!
- 16. Demo Time!
- 17. Q&A
- 18. If I push changes from my local files to my GitHub repo, do they automatically get pulled into Rasa X?
- 19. Can different Rasa X users push changes to their own branches?
- 20. When I launch Rasa X, I don’t see an option to enable experimental features. What could cause this?
- 21. Resources
- 22. ● Installing Rasa X on a server ● Connecting your assistant using Integrated Version Control ● Common workflows with Rasa Open Source and Rasa X Check out Episodes 9 and 10 of the Rasa Masterclass
- 23. ● Integrated Version Control: Linking Rasa X with Git-based Development Workflows (Rasa blog) ● Model testing and CI for conversational software (Rasa blog) ● Integrated Version Control Installation and Setup (docs) ● Improving the Assistant - Integrated Version Control (docs) ● Masterclass Ep #9 - Improving the assistant: Setting up the Rasa X (YouTube) ● How Rasa’s Research Team Uses Rasa X to Build Carbon Bot (YouTube) Further Reading
- 25. Ty Dunn Product Manager t.dunn@rasa.com Rasa Technologies GmbH, Schönhauser Allee 175, 10119 Berlin, Germany Get in touch!