Open jwfromm opened 1 month ago
Name | Link |
---|---|
Latest commit | ae2305dbabddd4f5d60d3659e1d48ffc6f009b93 |
Latest deploy log | https://app.netlify.com/sites/pytorch-fbgemm-docs/deploys/66b503e3a175200008bb6ee6 |
Deploy Preview | https://deploy-preview-2951--pytorch-fbgemm-docs.netlify.app |
Preview on mobile | Toggle QR Code...Use your smartphone camera to open QR code link. |
To edit notification comments on pull requests, go to your Netlify site configuration.
This pull request was exported from Phabricator. Differential Revision: D60972273
Summary: For performanc evaluation purposes, it may be convenient to specify a specific kernel be used for a given workload. This diff makes it easy to do so from python by adding the new
kernel_name
argument to fp8_rowwise.When
kernel_name
is provided, we use a lookup table to grab the specific kernel. When its not provided, we use our usual dispatch and heuristics.Differential Revision: D60972273