avinashcpandey comments

Results 5 comments of


                                            avinashcpandey

LRN has not been registered, ConvBatchnormScaleReluPool has not been registered error with Anakin inference engine

Thanks for the update! Can you check one issue(FPE error) i raised with Anakin(AMD GPU backend) ? https://github.com/PaddlePaddle/Anakin/issues/516

Query: Is running two instances of MLPerf valid for Offline and Server Submission

Thanks @arjunsuresh and @saar-eliad for the prompt response. I will explore as you suggested.

Feasibility of Adding a New Backend Library to oneDNN

Thanks @vpirogov for the prompt response! **Few follow up question:** 1. In workloads like FlashAttention, each thread executes many tiny GEMMs, and the current primitive-based API introduces overheads (descriptor creation,...

Feasibility of Adding a New Backend Library to oneDNN

Thank you @vpirogov for sharing all this detailed information—it’s extremely helpful for us in shaping the solution. We plan to pursue both paths in parallel: **1. Adding ZenDNN as a...

Feasibility of Adding a New Backend Library to oneDNN

Thank you @vpirogov for the additional information. The branch ZenDNN/zendnnl is expected to become the main branch this week. ZenDNN has recently undergone a strategic re-architecture and refactoring aimed at...