AndrewZhaoLuo opened a new pull request #9336: URL: https://github.com/apache/tvm/pull/9336
The actual topi assume IOHW kernel inputs. It seems for convs we are used to OIHW layouts but for some frameworks (e.g. pytorch) conv2d_transpose is done with IOHW kernel layouts. This has caused me a great deal of confusion. It seems like the current code only works because people mix up the layout convention many many times and it cancels out. This is my attempt at untying the knot and making it much more consistent. -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
