XNNPACK
XNNPACK copied to clipboard
google
High-efficiency floating-point neural network inference operators for mobile, server, and Web
![copybara-service[bot] avatar](https://avatars.githubusercontent.com/in/44061?v=4 "copybara-service[bot] avatar"){kind=avatar}
Add XNNPACK delegate per operator profiling to TFLite
Remove incorrect check on dimensions for split node's split dimension input tensor It is a 0-D tensor, the check was trying to check that it is a 1-D tensor. Also...
Rename quantized concatenation tests to include quantization in test name
[XNNPACK] Support SPLIT operator with 2 outputs
This PR adds support for building WASM version of XNNPACK with emcmake * New Build script `build-wasm.sh` to build XNNPACK with emscripten and emcmake * Options - `enable_tests`, `enable_benchmarks`, `enable_simd`...
Make SSE2 microkernels consistent with neon zip microkernels. - DEC is now MOV - Register usage is reduced for MOV - REUSE input pointers point to previous pointer instead of...