Make FloatToFloat16 conversion 75x faster using SVE2 instructions (#3… #8854
facebook-github-botbuild_wheels_linux_aarch64.yml
on: push
generate-matrix
/
generate
3s
Matrix: build
Artifacts
Produced during runtime
| Name | Size | |
|---|---|---|
|
pytorch_FBGEMM__3.10_cpu_aarch64
|
2.69 MB |
|
|
pytorch_FBGEMM__3.10_cu126_aarch64
|
446 MB |
|
|
pytorch_FBGEMM__3.11_cpu_aarch64
|
2.69 MB |
|
|
pytorch_FBGEMM__3.11_cu126_aarch64
|
446 MB |
|
|
pytorch_FBGEMM__3.12_cpu_aarch64
|
2.69 MB |
|
|
pytorch_FBGEMM__3.12_cu126_aarch64
|
446 MB |
|
|
pytorch_FBGEMM__3.13_cpu_aarch64
|
2.69 MB |
|
|
pytorch_FBGEMM__3.13_cu126_aarch64
|
446 MB |
|
|
pytorch_FBGEMM__3.9_cpu_aarch64
|
2.69 MB |
|
|
pytorch_FBGEMM__3.9_cu126_aarch64
|
446 MB |
|