mirror of https://github.com/llvm/torch-mlir
25a5a22cbd
Linalg has quantized specific operations. We can lower to these operations when there is a known zeropoint and scale operations. This allows the `convolution` to occur with lower bitwidth's, improving the overall performance. |
||
|---|---|---|
| .. | ||
| CAPI | ||
| Conversion | ||
| Dialect | ||
| RefBackend | ||
| CMakeLists.txt | ||
| InitAll.cpp | ||