Spaces:
Paused
Paused
Commit History
-
FORCE: HF sync issue - change README pinned + sync file v0.3.21
Peter Michael Gits7f07efb -
test: Re-test LlamaIndex with proper branch synchronization v0.3.15β
Peter Michael Gits +1: ClaudeNow that we fixed the branch sync issue, test if LlamaIndex works: - Switch app_file back to app.py (main ChatCal functionality) - Add back LlamaIndex with older stable versions - Keep pydantic>=2.7.0,<2.10.0 constraints - Test if the original Pydantic Secret import error is resolved This will determine if our dependency conflicts were real or just due to HF not seeing our updates. π€ Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>d1f39bc -
fix: Switch from Docker to Gradio SDK due to HF caching issues v0.3.14β
Peter Michael Gits +1: ClaudeHF Spaces is aggressively caching old Dockerfile despite multiple updates. Switch to reliable Gradio SDK approach: - README.md: sdk: gradio, app_file: app_simple.py - Created app_simple.py: Basic Gradio test without dependencies - Ultra-minimal requirements: only gradio + python-dotenv - This should work immediately without Docker build complexity Goal: Get basic functionality working, then add features incrementally. π€ Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>a44c7cd
-
feat: Switch to Docker deployment for environment consistency v0.3.8β
Peter Michael Gits +1: ClaudeBrilliant insight! Use Docker to replicate exact Google Cloud Run environment: - Add Dockerfile with Python 3.11 (matching Cloud Run) - Switch HF Space from gradio SDK to docker SDK - Use stable package versions that worked in Cloud Run - Eliminates Pydantic/LlamaIndex compatibility issues entirely - Same controlled environment as Cloud Run - Switch from debug_app.py back to main app.py This bypasses all dependency resolution conflicts by using the exact same environment that's working in production. π€ Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>2bdbfa0 -
fix: Change hardware to free cpu-basic tierβ
Peter Michael Gits +1: ClaudeSwitch from t4-small to cpu-basic to use free Hugging Face hardware. This should be sufficient for the voice-enabled ChatCal functionality. π€ Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>979da79 -
debug: Add debug app to identify crash causeβ
Peter Michael Gits +1: ClaudeTemporarily switch to debug_app.py to diagnose: - Import status of all dependencies - Environment variable configuration - Exact error location causing 500 error This will help identify if it's: - Pydantic/LlamaIndex import issues - Missing environment variables - Other dependency conflicts π€ Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>772c21a -
feat: Initial ChatCal Voice-Enabled Assistant v0.1.0β
Peter Michael Gits- Migrated ChatCal to Hugging Face Gradio platform - Added voice interaction capabilities with STT/TTS integration - Implemented session management for Gradio environment - Created LLM fallback chain: Groq β Anthropic β Mock - Built responsive voice-enabled chat interface - Added demo mode for development and testing - Security: Using HF environment variables for secrets π€π Generated with Claude Code32e756c