tkonolige opened a new pull request #8130: URL: https://github.com/apache/tvm/pull/8130
This flag is causes CUBLAS to use tensor cores on all operations. With f32 or f64 operations, this leads to loss of accuracy. #7995 Supersedes #8108. @areusch @comaniac -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
