Skip to content

CONFIG_CHECKLIST.md

Latest commit

 

History

History
261 lines (187 loc) · 6.61 KB

CONFIG_CHECKLIST.md

File metadata and controls

261 lines (187 loc) · 6.61 KB
Raw

✅ Aria Environment Configuration — Quick Checklist

Status: May 16, 2026 — Environment configured and ready Python: 3.14 (configured for main + all sub-projects)

📋 What Was Configured

✅ Main Aria Workspace

  • Location: /workspaces/Aria
  • Venv: /workspaces/Aria/.venv
  • Status: Ready for pip install

✅ Quantum ML Sub-Project

  • Location: ai-projects/quantum-ml
  • Venv: Shares main .venv
  • Key: qiskit==1.3.0, azure-quantum[qiskit]

✅ Chat CLI Sub-Project

  • Location: ai-projects/chat-cli
  • Venv: Shares main .venv
  • Key: openai>=1.58.0

✅ LoRA Training Sub-Project

  • Location: AI/microsoft_phi-silica-3.6_v1
  • Venv: Shares main .venv
  • Key: PyTorch training scripts

✅ Provider Detection Chain

  • LMStudio → Azure OpenAI → OpenAI → LoRA → Local fallback
  • Config: local.settings.json (all providers optional)

🚀 Next Steps (Copy & Paste Ready)

1️⃣ Install All Dependencies

cd /workspaces/Aria

# Main
pip install -r requirements.txt

# Optional: Dev tools (linting, testing)
pip install -r requirements-dev.txt

# Sub-projects
pip install -r ai-projects/quantum-ml/requirements.txt
pip install -r ai-projects/chat-cli/requirements.txt
pip install -r ai-projects/llm-maker/requirements.txt
pip install -r ai-projects/cooking-ai/requirements.txt

2️⃣ Quick Validation

# Fast validation (all checks pass? → good to go)
python scripts/fast_validate.py

# Critical deps check
python3 -c "
import sys
pkgs = ['torch', 'transformers', 'qiskit', 'azure-quantum', 'openai']
missing = [p for p in pkgs if __import__(p.replace('-', '_'), fromlist=['']) is None]
print('✓ All critical packages' if not missing else f'✗ Missing: {missing}')
sys.exit(0 if not missing else 1)
"

3️⃣ Configure Secrets (Optional)

If you have API keys, edit local.settings.json:

{
    "Values": {
        "AZURE_OPENAI_API_KEY": "your-key",
        "AZURE_OPENAI_ENDPOINT": "your-endpoint",
        "OPENAI_API_KEY": "your-key",
        "LMSTUDIO_BASE_URL": "http://localhost:1234/v1"
    }
}

4️⃣ Start Azure Functions

func host start

5️⃣ Test API Endpoints (In another terminal)

# Health check
curl http://localhost:7071/api/ai/status | python3 -m json.tool

# Provider probe
curl -X POST http://localhost:7071/api/ai/provider-probe \
  -H 'Content-Type: application/json' \
  -d '{"provider":"auto"}' | python3 -m json.tool

# List routes
curl http://localhost:7071/api/ai/routes | python3 -m json.tool

📦 Dependency Summary

Category Key Package Min Version
AI/Chat openai 1.58.0
Quantum qiskit 1.3.0
Quantum ML qiskit-machine-learning 0.8.2
Azure Quantum azure-quantum 1.0.0
Deep Learning torch 2.8.0
Transformers transformers Latest
Fine-tuning peft Latest
Functions azure-functions Latest
Database sqlalchemy 2.0.36
Cosmos azure-cosmos 4.7.0
Telemetry azure-monitor-opentelemetry 1.0.0

🔌 Provider Detection (Auto-Detected Order)

1. LMStudio?     → LMSTUDIO_BASE_URL set?
2. Azure OpenAI? → All 4 vars set?
3. OpenAI?       → OPENAI_API_KEY set?
4. LoRA?         → Explicit --provider lora?
5. Local?        → Zero-config fallback ✓

To force a provider:

# CLI
python ai-projects/chat-cli/src/chat_cli.py --provider openai --once "Hello"

# API (via function_app.py)
curl -X POST http://localhost:7071/api/chat \
  -H 'Content-Type: application/json' \
  -d '{"provider":"azure-openai","message":"Hello"}'

✅ Validation Commands

Test Each Component

# Chat CLI (local, no setup needed)
python ai-projects/chat-cli/src/chat_cli.py --provider local --once "Test"

# Quantum simulation
python -c "from qiskit import QuantumCircuit; print('✓ qiskit')"

# Torch GPU (if available)
python -c "import torch; print(f'✓ torch (GPU: {torch.cuda.is_available()})')"

# Azure Functions
func host start  # Then: curl http://localhost:7071/api/ai/status

Run Test Suite

# Unit tests only
pytest tests/ -m "not slow and not azure" -v

# With coverage
pytest tests/ -m "not slow and not azure" --cov

# Using test runner
python scripts/test_runner.py --unit

🛠️ Optional: Isolated Venvs

If you need separate venvs per project (e.g., conflicting deps), create them:

# Quantum ML isolated
cd ai-projects/quantum-ml
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate
pip install -r requirements.txt

# Chat CLI isolated
cd ai-projects/chat-cli
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt

# Update VS Code settings → python.defaultInterpreterPath

📚 Full Documentation

See ENVIRONMENT_SETUP.md for:

  • ✓ Detailed dependency lists
  • ✓ Provider configuration examples
  • ✓ API endpoint docs
  • ✓ Troubleshooting
  • ✓ Advanced setup (SQL, Cosmos, Telemetry)

🆘 Quick Troubleshooting

Problem Fix
ModuleNotFoundError: torch pip install torch>=2.8.0
ModuleNotFoundError: qiskit pip install qiskit==1.3.0
ModuleNotFoundError: openai pip install openai>=1.58.0
Provider returns "local" Check /api/ai/status for env vars
/api/chat returns 500 Run /api/ai/provider-probe to diagnose
Functions fail to start pip install azure-functions

📝 Files Modified

✅ Created: ENVIRONMENT_SETUP.md (full reference guide) ✅ Referenced: local.settings.json.example (provider config template) ✅ Noted: requirements.txt (main + all sub-projects) ✅ Ready: pyproject.toml (project metadata)

✨ You're Ready To

  • ✅ Install all Python dependencies
  • ✅ Start Azure Functions runtime
  • ✅ Test chat with multiple providers
  • ✅ Run quantum simulations
  • ✅ Fine-tune LoRA models
  • ✅ Execute integrated tests

Next: Run the 4 quick install commands above! 🎉

Generated: May 16, 2026 Configuration: ✅ Complete Ready: ✅ Yes