MPSCNNSoftMaxNode's result image format?!

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?

Up vote post of Hardy Down vote post of Hardy
475 views
Posted by
Hardy
Reply
Add a Comment

Accepted Reply

I am going to answer my own question: when using graphs (of MPSNNGraph type), setting the graph's format to MPSImageFeatureChannelFormatFloat32 or to any other suitable type does the job.

Posted by
Hardy
Up vote reply of Hardy Down vote reply of Hardy
Post marked as solved
Add a Comment

Replies

I am going to answer my own question: when using graphs (of MPSNNGraph type), setting the graph's format to MPSImageFeatureChannelFormatFloat32 or to any other suitable type does the job.

Posted by
Hardy
Up vote reply of Hardy Down vote reply of Hardy
Post marked as solved
Add a Comment