Spaces:
Sleeping
Sleeping
Commit History
-
requirements.txt 복구 (HuggingFace Spaces용)
Yoon-gu Hwang +1: Claude- 이전 버전의 requirements.txt 복구 - HuggingFace Spaces Docker 이미지 빌드용 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>a71d082 -
런타임 에러 수정: langchain-core 의존성 추가
Yoon-gu Hwang +1: Claude- pyproject.toml에 langchain-core 명시적 추가 - ModuleNotFoundError: No module named 'langchain_core' 해결 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>a0ecc57 -
Basic Chatbot 제거 및 ML Pipeline만 남김
Yoon-gu Hwang +1: Claude- basic_chatbot_workflow.py 삭제 - app.py를 ML Pipeline 단일 인터페이스로 간소화 - TabbedInterface 제거하고 단일 ChatInterface로 변경 - README 업데이트 (Basic Chatbot 기능 설명 제거) 더 집중된 ML Pipeline 전용 애플리케이션 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>820a28c -
프로젝트 구조 평탄화 및 불필요한 파일 정리
Yoon-gu Hwang +1: Claude- 폴더 구조 제거하고 모든 Python 파일을 최상위로 이동 - basic_chatbot/workflow.py → basic_chatbot_workflow.py - ml_pipeline/workflow.py → ml_pipeline_workflow.py - basic_chatbot/, ml_pipeline/ 폴더 삭제 - tab.py 파일들 삭제 (app.py에서 직접 구현) - requirements.txt 삭제 (pyproject.toml 사용) - README 업데이트 더 간단하고 명확한 프로젝트 구조 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>5dc0c74 -
불필요한 파일 정리
Yoon-gu Hwang +1: Claude- basic_chatbot/tab.py 삭제 (app.py에서 직접 함수 정의) - ml_pipeline/tab.py 삭제 (app.py에서 직접 함수 정의) - requirements.txt 삭제 (pyproject.toml과 uv.lock으로 관리) - __pycache__ 디렉토리들 정리 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>0ffbf55 -
LangSmith observability 추가 및 UI 개선
Yoon-gu Hwang +1: Claude- LangSmith 추적 기능 추가 (observability) - .env에 LANGCHAIN_TRACING_V2, LANGCHAIN_API_KEY 설정 - .env.example 템플릿 파일 생성 - pyproject.toml에 langsmith 의존성 추가 - README에 설정 방법 상세 문서화 - Chatbot UI 기능 개선 - Like/Dislike 버튼 추가 (gr.Blocks로 구조 변경) - Copy, Copy All, Share 버튼 추가 - Clear 버튼 기본 포함 - ML Pipeline 예제 확장 - 전체 파이프라인 예제 (4단계) - 조합 예제 추가 (2-3단계 조합) - 단일 기능 예제 유지 - 다양한 시나리오 테스트 가능 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>3b42a60 -
Supervisor 프롬프트 개선: 요청된 작업만 수행하도록 수정
Yoon-gu Hwang +1: Claude- 강제된 순차 실행 지시 제거 - 사용자 요청 범위 내에서만 작업하도록 명확화 - 불필요한 다음 단계 자동 진행 방지 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>c6a59e0 -
UI 제목 정리 및 ML Pipeline 예제 확장
Yoon-gu Hwang +1: Claude변경 사항: 1. 제목에서 "Nested Thoughts" 용어 제거 - "Multi-Agent Chatbot (Nested Thoughts)" → "Multi-Agent Chatbot" - "ML Pipeline Automation (Nested Thoughts)" → "ML Pipeline Automation" - "Multi-Agent Systems with Nested Thoughts" → "Multi-Agent Systems" 2. ML Pipeline 예제 추가 (총 6개) - 기존: 전체 파이프라인 예제 3개 - 추가: 개별 단계 예제 3개 • 사전학습만 하기 • 파인튜닝만 하기 • 평가만 하기 사용자가 원하는 특정 단계만 실행할 수 있도록 개선 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>6097ee9 -
불필요한 테스트 및 디버깅 파일 정리
Yoon-gu Hwang +1: Claude삭제된 파일: - app_old.py (백업 파일) - debug_stream.py (디버깅용) - test_agent_content.py (테스트용) - test_agent_status.py (테스트용) - test_nested_structure.py (테스트용) - test_pending_status.py (테스트용) HuggingFace Spaces 구동에 필요한 파일만 유지 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>b8f2202 -
Nested Thought UI를 메인 앱으로 교체
Yoon-gu Hwang +1: Claude변경 사항: - app_nested_thought.py를 app.py로 교체 - 기존 app.py는 app_old.py로 백업 - 탭 순서를 ML Pipeline → Basic Chatbot으로 변경 (기존 app.py와 동일) - ssr_mode=False 설정 추가 이제 HuggingFace Spaces에서 Nested Thought 기능이 적용된 UI가 기본으로 표시됩니다. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>7e711f2 -
Nested Thought 구조를 활용한 Multi-Agent Gradio 앱 구현
Yoon-gu Hwang +1: Claude주요 기능: - Gradio의 Nested Thought 패턴을 LangGraph multiagent 시스템에 적용 - Agent와 tool 호출의 계층적 시각화 (parent_id 기반) - 실시간 작업 진행 상황 표시 (pending → done 전환) - Agent header에 현재 실행 중인 작업 내용 자동 업데이트 - 최종 답변은 metadata 없는 일반 메시지로 표시 구현 내용: - Basic Chatbot (Research + Math agents) - ML Pipeline (Data extraction → Pretraining → Finetuning → Evaluation) - 각 agent의 서브 작업들이 nested 구조로 실시간 표시 - Sub node들이 작업 중일 때 pending 상태로 UI에 표시 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>1a5d960
-
Reorganize code structure by workflow
Yoon-gu Hwang +1: Claude- Create workflow-based folder structure (basic_chatbot, ml_pipeline) - Move workflow and tab files into respective folders - Update import paths to use relative imports - Add __init__.py files for Python module structure - Improve ML pipeline examples with realistic Korean use cases 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>4765d13 -
Rename graph.py to basic_workflow.py for consistency
Yoon-gu Hwang +1: Claude- Rename graph.py to basic_workflow.py to match ml_workflow.py naming - Update import in tab_basic_chatbot.py - Update README.md project structure section Both workflow files now have equal, consistent naming: - basic_workflow.py: Research and math agents - ml_workflow.py: ML pipeline agents 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>562380b -
Refactor: Move tools into workflow files and translate to Korean
Yoon-gu Hwang +1: Claude- Move all ML pipeline tools from tools.py into ml_workflow.py - Delete tools.py to consolidate code structure - Translate all tool docstrings to Korean in ml_workflow.py - Translate all tool docstrings and agent prompts to Korean in graph.py - Reorder tabs in app.py: ML Pipeline is now the first tab Both graph.py and ml_workflow.py now have equal structure with their own tools and Korean language support. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>f26f95b -
Add ML pipeline automation system with tabbed interface
Yoon-gu Hwang +1: Claude- Create tabbed interface with Basic Chatbot and ML Pipeline tabs - Add tab_basic_chatbot.py for existing multi-agent chatbot - Add tab_ml_pipeline.py for ML pipeline automation - Add ml_workflow.py with 4 specialized agents (Korean prompts): - data_extraction_expert: SQL and RDB operations - pretraining_expert: Data preparation and model pretraining - finetuning_expert: Finetuning data creation and classification training - evaluation_expert: Model evaluation with precision, recall, F1, accuracy - Add tools.py with realistic ML pipeline tool functions: - extract_events_from_rdb, prepare_pretraining_data - pretrain_model, create_finetuning_data - train_classification_model, evaluate_model - Update README.md with new features and project structure - Simplify app.py to use TabbedInterface 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>a8500b3 -
Fix HuggingFace Spaces startup error
Yoon-gu Hwang +1: Claude- Disable examples caching to prevent startup validation errors - Add HuggingFace Spaces deployment instructions to README - Add OPENAI_API_KEY secret configuration guide This fixes the pydantic validation error that occurred when Gradio tried to cache examples during startup on HuggingFace Spaces. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>c9a0817 -
Update README.md with gradio version and documentation
Yoon-gu Hwang +1: Claude- Update sdk_version from 5.37.0 to 5.49.1 - Add comprehensive project description and features - Add setup instructions with uv commands - Add project structure and technologies sections - Improve documentation for new users 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>b8c099d -
Add uv project configuration and fix dependencies
Yoon-gu Hwang +1: Claude- Add pyproject.toml for uv package management with all required dependencies - Add .gitignore to exclude .env, __pycache__, and virtual environments - Add python-dotenv for environment variable management - Remove unused langchain_teddynote.messages import from app.py - Load .env file automatically in graph.py for OPENAI_API_KEY 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>2b02283
-
Update requirements.txt
Verified signatureb5ddcd8
-
Update README.md
Verified signature6027e50 -
Update app.py
Verified signaturef647d10 -
Update app.py
Verified signature757bf39 -
Update app.py
Verified signature72ce352 -
Update app.py
Verified signatureb6ebb96 -
Create graph.py
Verified signature4f3b1c9 -
Create app.py
Verified signature9e93462 -
Create requirements.txt
Verified signature964ccd2 -
initial commit
Verified signature055b0c7