Hi all,
I was compiling the TVM Tensorflow tutorial (https://docs.tvm.ai/tutorials/frontend/from_tensorflow.html) when I’ve received many many occurrences of the above noted warning. The applications runs without a problem. I was just curious what this warning is caused by and whether or not it can lead to critical problems on the long run. I’m using TVM commit a6c42b34782059b82c56ac53a9df63d55e452120 (stable from the master) with Tensorflow 1.14 on Ubuntu 18.04. Explanations & ideas are much appreciated!
Here a couple of warning lines:
WARNING:root:Attribute T is ignored in relay.sym.conv2d
WARNING:root:Attribute explicit_paddings is ignored in relay.sym.conv2d
WARNING:root:Attribute use_cudnn_on_gpu is ignored in relay.sym.conv2d
WARNING:root:Attribute _node_name is ignored in relay.sym.conv2d
WARNING:root:Attribute _target_layout is ignored in relay.sym.conv2d
WARNING:root:Attribute _input_shapes is ignored in relay.sym.conv2d
WARNING:root:Attribute T is ignored in relay.sym.batch_norm
WARNING:root:Attribute _output_shapes is ignored in relay.sym.batch_norm
Best regards, Robert