Pinned Loading

1. CUDA-Accelerated-LiDAR-PointPillars-Preprocessing CUDA-Accelerated-LiDAR-PointPillars-Preprocessing Public

   C++/CUDA PointPillars LiDAR preprocessing pipeline with KITTI loading, pillar scatter, BEV pseudo-image generation, tests, and CUDA range filtering.

   Cuda 3

2. GEMM-optimization GEMM-optimization Public

   CUDA FP32 GEMM optimization with loop unrolling, shared memory tiling, register tiling, benchmarking, and Nsight profiling.

   Cuda 1

3. mini-vllm-cuda mini-vllm-cuda Public

   CUDA kernels for LLM decode-stage inference, built as a PyTorch extension with correctness tests and latency benchmarks.

   Python