draft: batch/simulation fixups #890
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Consolidate hotel-api fixtures into a single hotel module with an integrated service layer. Rename duet_cli.py to hotel_simulation.py and personalities.json to personas.json for clarity.
Introduce a flexible adapter system for chat response transformations. Includes ChatAdapter protocol, AdapterPipeline for chaining adapters, and built-in adapters for common transformations like JSON extraction and format detection.
Improve the core agent simulation module with better conversation handling, enhanced results tracking, scenario grouping support, and improved display utilities. Refactor simulation execution for better modularity and testability.
Introduce a comprehensive checkers system for validating agent simulation outputs. Includes LLM-based checkers for evaluating task completion, response quality, and conversation flow.
Restructure the metrics module with improved collectors, enhanced built-in metrics, and add DeepEval integration for comprehensive evaluation metrics including conversation coherence and task completion assessment.
Introduce MemoryTraceHandler and TraceAnalyzer for capturing and analyzing LLM calls, tool invocations, and token usage during simulation runs. Extract tool calls and usage from traces for more reliable data collection.
Introduce a comprehensive storage layer with key-value and SQL store abstractions. Add connection managers for SQLite and PostgreSQL databases, along with a memory-based trace handler for capturing audit traces.
Introduce a store abstraction for persisting evaluation data including scenarios, personas, simulation runs, and results. Includes file-based and key-value store implementations for flexible data persistence.
Introduce a FastAPI-based evaluation API for managing simulation scenarios, personas, and runs. Add ExecutionManager for orchestrating simulation execution with real-time progress updates via SSE.
Introduce a comprehensive evaluation dashboard built with React and TypeScript. Features include scenario management, persona configuration, simulation run execution with real-time progress, and detailed results visualization with conversation views and metrics display.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.