export efficientnet_b0b model failed

[waquey]

Hello, I've tried to export efficientnet_b0b. However, it failed.

The following is my code segment:

net = glcv2_get_model("efficientnet_b0b", pretrained=True)
net.hybridize()

x = np.zeros([1,3,224,244])
x = mx.nd.array(x)

net.forward(x)
net.export('efficientnet_b0b')

And the error message is as following:

File "/usr/local/lib/python2.7/dist-packages/mxnet/gluon/block.py", line 915, in forward
return self._call_cached_op(x, *args)
File "/usr/local/lib/python2.7/dist-packages/mxnet/gluon/block.py", line 805, in _call_cached_op
self._build_cache(*args)
File "/usr/local/lib/python2.7/dist-packages/mxnet/gluon/block.py", line 757, in _build_cache
data, out = self._get_graph(*args)
File "/usr/local/lib/python2.7/dist-packages/mxnet/gluon/block.py", line 749, in _get_graph
out = self.hybrid_forward(symbol, *grouped_inputs, **params) # pylint: disable=no-value-for-parameter
File "/usr/local/lib/python2.7/dist-packages/gluoncv2/models/efficientnet.py", line 393, in hybrid_forward
x = self.features(x)
File "/usr/local/lib/python2.7/dist-packages/mxnet/gluon/block.py", line 548, in call
out = self.forward(*args)
File "/usr/local/lib/python2.7/dist-packages/mxnet/gluon/block.py", line 932, in forward
return self.hybrid_forward(symbol, x, *args, **params)
File "/usr/local/lib/python2.7/dist-packages/mxnet/gluon/nn/basic_layers.py", line 117, in hybrid_forward
x = block(x)
File "/usr/local/lib/python2.7/dist-packages/mxnet/gluon/block.py", line 548, in call
out = self.forward(*args)
File "/usr/local/lib/python2.7/dist-packages/mxnet/gluon/block.py", line 932, in forward
return self.hybrid_forward(symbol, x, *args, **params)
File "/usr/local/lib/python2.7/dist-packages/gluoncv2/models/efficientnet.py", line 274, in hybrid_forward
x = F.pad(x, mode="constant", pad_width=calc_tf_padding(x, kernel_size=3, strides=2), constant_value=0)
File "/usr/local/lib/python2.7/dist-packages/gluoncv2/models/efficientnet.py", line 43, in calc_tf_padding
height, width = x.shape[2:]
AttributeError: 'Symbol' object has no attribute 'shape'

Do you know how to modify it? Thanks

[osmr]

Hi, yes, the current implementation of the TF-like EfficientNet models (models with b-suffix) does not allow hybridization. EfficientNet models without b-suffix don't have such a problem.

For the EfficientNet-b model, you need to exclude the calculation of x.shape in the calc_tf_padding function. This requires careful modification of the model script (the weights will be preserved.) I can’t do this right now due to lack of time.

The standard way in this situation is to analyze what values are obtained in these places during the debug of a non-hybridized network (they will be different depending on the size of the network input!) And pass the valid values of height and width to calc_tf_padding function.