1 Reply
      Latest reply on Mar 22, 2020 5:15 AM by Hardy
      Hardy Level 1 Level 1 (0 points)

        I am using a neural network with a data type MPSDataTypeFloat32 but it seems to be that the final MPSCNNSoftMaxNode for the inference run  returns an image using MPSDataTypeFloat16. Is this always the case? How do the nodes determine the output data type? Can I change them?