masahi commented on issue #11780: URL: https://github.com/apache/tvm/issues/11780#issuecomment-1159826773
Does it compile and run without autotvm? I think maybe the ONNX frontend is generating a conv2d op whose input width and height are dynamic, when they shouldn't be. I've seen the same issue when for MaskRCNN when imported via ONNX. If that's the case, this is not an autotvm problem. -- 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]
