Muyue/activity-tracker
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
13 Commits 1 Branch 0 Tags
Commit Graph

3 Commits

Author SHA1 Message Date
Augustin
e17a4dd9d0 Feature: Add ONNX model support with NPU/DirectML acceleration 
- Replace GGUF models with ONNX models optimized for DirectML
- Add Microsoft Phi-3 Mini DirectML (INT4, 2.4GB)
- Add Xenova ONNX models (DistilBERT, BERT, MiniLM, CLIP)
- Update model catalog with working HuggingFace URLs
- Create ONNX/NPU integration test suite (tests/onnx_npu_test.rs)
- Successfully test DistilBERT ONNX loading with DirectML
- Verify NPU session creation and model inputs/outputs

Test Results:
- ✅ NPU Detection: Intel AI Boost NPU (via DirectML)
- ✅ ONNX Session: Created successfully with DirectML
- ✅ Model: DistilBERT (268 MB) loaded
- ✅ Inputs: input_ids, attention_mask
- ✅ Output: logits
- ⚡ Performance: Ready for NPU hardware acceleration

All tests passing with NPU-accelerated ONNX inference

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-10-16 18:53:52 +02:00
Augustin
c25711dd1e Feature: Add NPU inference tests and model download capability 
- Add comprehensive NPU inference performance tests (tests/npu_inference_test.rs)
  - NPU session creation validation
  - DirectML configuration verification
  - Classifier NPU integration testing
  - Performance baseline: 21,190 classifications/sec
- Implement HTTP-based model download using ureq (src/ai/models.rs)
  - Progress tracking during download
  - Chunk-based file writing
  - Error handling for network failures
- Update CLI model management commands (src/main.rs)
  - Enhanced model listing with download status
  - Improved error messages for unknown models
- Add ureq dependency for HTTP downloads (Cargo.toml)

All 39 tests passing (30 unit + 5 AI integration + 4 NPU inference)

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-10-16 14:50:40 +02:00
Augustin
03950aafca Test: Add AI/NPU integration tests 
Tests:
- NPU detection (Intel AI Boost)
- DirectML availability check
- Classifier creation and initialization
- Rule-based classification fallback
- Multiple classification scenarios

Test Results:
✅ NPU Device: Intel AI Boost NPU (via DirectML)
✅ NPU Available: true
✅ DirectML: Enabled
✅ Classification accuracy: 100%
✅ All 5 integration tests passing

Test cases verified:
- VSCode -> Development (confidence: 1.0)
- Chrome -> Research (confidence: 1.0)
- Zoom -> Meeting (confidence: 0.95)
- Figma -> Design (confidence: 1.0)
- Terminal -> Development (confidence: 0.8)

Total tests: 35/35 passing

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-10-16 14:39:32 +02:00