Fetching metadata from the HF Docker repository...
-
Fix Dockerfile heatmap risk level
-
Fix AMD Readiness Score and WMMA Heatmap
-
feat: AST-level Python analysis + MI300X benchmark suite
- AST Transformer: Uses Python's ast module for tree-level analysis
- Detects device assignments, CUDA call sites, embedded kernels
- Identifies env mutations and inline kernel strings structurally
- Not regex — real compiler-level understanding of code structure
- MI300X Benchmark Suite: Ready to run on AMD Developer Cloud
- Device info, memory bandwidth, and GEMM TFLOPS tests
- Saves JSON results as proof of AMD GPU usage
- Wired AST findings into analyzer, API, and orchestrator -
feat: 9-agent pipeline with hardware-aware analysis, drift detection, build copilot
- Hardware-Aware Scanner: Warp→Wavefront, TensorCore→MFMA intrinsic lowering
- Exploration Scanner: Curiosity-driven implicit CUDA assumption detection
- Verification Pass: Rescue branches for leftover CUDA artifacts
- Health Monitor: Per-line saliency map + migration drift detection
- Build Error Copilot: ROCm build error runbook with preemptive fixes
- WMMA Tensor Core sample: Hardest migration case demo
- 9-agent pipeline (up from 6), updated architecture diagram
-
Initial commit for AMD Developer Hackathon: Multi-Agent ROCm Migration Engine