vvchernov opened a new pull request, #13322: URL: https://github.com/apache/tvm/pull/13322
QLinearMatMul supports rank =2 only for both input tensors. It was extended using _qnn.op.dense and _qnn.op.batch_matmul for all ranks There are questions related to the following: 1. Different types of input tensors (int8 and uint8) 2. Output scaling factor and zero point. Just now it is x_scale*w_scale and 0, correspondingly. -- 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]
