spark-deep-learning icon indicating copy to clipboard operation
spark-deep-learning copied to clipboard

Output multiple tensors from TFImageTransformer

Open gyscos opened this issue 5 years ago • 0 comments

Hi! From a dataframe of images, how could I run a detection model, and retrieve in one pass both the detection boxes and the detection classes?

I can currently use either outputTensor="detection_classes:0" or outputTensor="detection_boxes:0" when defining the TFImageTransformer, but I can't seem to output both, without running the transformer twice.

Is there a way to output multiple tensors from a single TFImageTransformer?

gyscos avatar Dec 14 '18 23:12 gyscos