[GitHub] [tvm] junluan opened a new issue #8299: [BUG][Relay][QNN] relay.qnn.op.conv2d failed with group > 1 and weight scale's length is [out_c]

[GitBox <gi...@apache.org>]

junluan opened a new issue #8299:
URL: https://github.com/apache/tvm/issues/8299


   For example:
   
   When the input tensor with shape [1, 256, 56, 56], weight tensor with shape [256, 8, 3, 3], weight scale with shape [256], group is 32, then this can be failed because of this check: [https://github.com/apache/tvm/blob/main/src/relay/qnn/op/convolution.cc#L81](https://github.com/apache/tvm/blob/main/src/relay/qnn/op/convolution.cc#L81)
   
   This check needs weight scale's length is the same as `out_c * in_c / group`. But when I use a quantized pytorch pretrained model, such as  `torchvision.models.quantization.resnext101_32x8d`, it's weight scale's length is [out_c]. I changed this requirement to out_c, it works fine.
   
   Thank you.


-- 
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:
users@infra.apache.org

[GitHub] [tvm] masahi commented on issue #8299: [BUG][Relay][QNN] relay.qnn.op.conv2d failed with group > 1 and weight scale's length is [out_c]

[GitBox <gi...@apache.org>]

masahi commented on issue #8299:
URL: https://github.com/apache/tvm/issues/8299#issuecomment-866522280


   Sounds like the same problem in https://github.com/apache/tvm/issues/7878?
   
   Pinging @anijain2305 for the bug fix.


-- 
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:
users@infra.apache.org