1 repository
Defines a matrix-matrix convolution computation with configurable dimensions, data types, and optional bias.
Distinct from Convolutional Kernel Optimizations: Distinct from Convolutional Kernel Optimizations: focuses on defining the convolution kernel, not optimizing it.
Explore 1 awesome GitHub repository matching artificial intelligence & ml · Convolution Kernel Configurators. Refine with filters or upvote what's useful.
TileLang is a Python-embedded domain-specific language compiler that JIT-compiles and autotunes GPU kernels. It uses a tile-based DSL, automatic software pipelining, and parallel autotuning to generate optimized GPU kernels at runtime. It supports tensor core operations with Pythonic syntax, automatic memory management, and thread mapping. The compiler searches over tile sizes, thread counts, and scheduling policies, compiling and benchmarking candidates in parallel to find the fastest kernel. It also caches compiled binaries and tuning results to disk for reuse across sessions. TileLang inc
Defines matrix-matrix convolution computations with configurable dimensions, data types, and optional bias.