I downloaded mobilenetv2 model from deeplab model zoo :
https://github.com/tensorflow/models/blob/master/research/deeplab/g3doc/...
And when I run the command on SNPE 1.14.0
"snpe-tensorflow-to-dlc --graph mobilenetv2_deeplab/frozen_inference_graph.pb --input_dim sub_7 513,513,3 --dlc mobilenetv2_deeplab.dlc --out logits/semantic/BiasAdd",
and I got th error as below:
ERROR - Encountered Error: non-broadcastable output operand with shape (65,65,64) doesn't match the broadcast shape (1,65,65,64)
Then I found the bug in eltiwise.py, line 127:" broadcasted_tensor += tensor" caused this error, I think author want to reshape tensor shape to the input "shape" of "def _broadcast_tensor(self, tensor, shape)", but the implemention is not right, I modified "broadcasted_tensor += tensor" to "broadcasted_tensor = broadcasted_tensor + tensor
broadcasted_tensor = broadcasted_tensor.reshape(shape)".
Please confirm this modification is correct.
Hi FangLiu,
The input dimensions provided for arguments are incorrect. Please find the below command for right usage,