Tree based new negative questions#19
Open
alexashalenina wants to merge 30 commits into
Open
Conversation
Co-authored-by: Anton Sinitsin <ant.sinitsin@gmail.com> Co-authored-by: Anton Sinitsin <30695750+xtinkt@users.noreply.github.com> Co-authored-by: Alexandra Olenina <olenina.alexandra.a@gmail.com> Co-authored-by: goodevening13 <aashutova12@yandex.ru> Co-authored-by: Vinogradov Ivan <vinogradov.ig@phystech.edu> Co-authored-by: Alina Shutova <91018099+goodevening13@users.noreply.github.com>
- Replace benchmark/state_tracking/ with unified benchmark runner at benchmark/ root - benchmark.py: supports mem0 RAG, mem0 Agent, mem-agent with model_profiles - Config uses OPENAI_API_KEY + OPENAI_BASE_URL (env vars, any OpenAI-compatible provider) - Data: 42 state_machine_location scenarios (0-5 transitions, 7 cases each) - Judge: configurable model via JUDGE_MODEL env var - Security: remove hardcoded API key from mem-agent/agent/settings.py - Remove .idea/ from tracking, update .gitignore for runtime artifacts - Add .env.example Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Separate config into clear sections: memory system definitions (mem0, mem_agent) and experiments with runtime params and model overrides. Prompts moved into self-contained dataset.yaml per data directory. - Add config_loader.py with Dataset/Experiment dataclasses - Refactor benchmark.py: run_experiment() replaces run_benchmark_for_model() - Experiments override model/api_key/base_url from memory system defaults - Judge updated for nested experiment output directories - Prompts relocated to data/state_machine_location/prompts/ - Add README with how-to guides for new datasets and experiments Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Default config now runs each memory system with a different model: - mem0 RAG → gpt-4o (OpenAI) - mem0 Agent (FC) → gemini-3-flash-preview (via LLM_PROVIDER) - mem-agent → gemini-3-pro-preview (via LLM_PROVIDER) Uses LLM_PROVIDER_API_KEY / LLM_PROVIDER_BASE_URL for non-OpenAI models. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Move benchmark.py, config files, judge/ and prompts/ from benchmark/ to repository root. Keep data directories in benchmark/. Fix .env and path resolution in all scripts. Merge benchmark/README.md into root README. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Benchmark migration
Adding account bench
… redundant files, wrote dataset.yaml
o-authored-by: Anton Sinitsin <ant.sinitsin@gmail.com> Co-authored-by: Anton Sinitsin <30695750+xtinkt@users.noreply.github.com> Co-authored-by: Alexandra Olenina <olenina.alexandra.a@gmail.com> Co-authored-by: goodevening13 <aashutova12@yandex.ru> Co-authored-by: Vinogradov Ivan <vinogradov.ig@phystech.edu> Co-authored-by: Alina Shutova <91018099+goodevening13@users.noreply.github.com>
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.