TypeError: <lambda>() got an unexpected keyword argument 'dtype'

[thima2017]

I'm trying to run stylize.py on Ubuntu 20.04 with the latest tensorflow (2.3.0) and keras (2.4.3). Tensorflow 1.* isn't available for python 3.8 shipped with Ubuntu 20.04.
I've changed tf.get_default_graph() to tf.compat.v1.get_default_graph() and tf.placeholder_with_default to tf.compat.v1.placeholder_with_default and everything seems worked until this:

Traceback (most recent call last):
  File "stylize.py", line 126, in <module>
    main()
  File "stylize.py", line 46, in main
    wct_model = WCT(checkpoints=args.checkpoints, 
  File "/full/path/WCT-TF2.3/wct.py", line 33, in __init__
    self.model = WCTModel(mode='test', relu_targets=relu_targets, vgg_path=vgg_path,
  File "/full/path/WCT-TF2.3/model.py", line 62, in __init__
    self.vgg_model = vgg_from_t7(vgg_path, target_layer=deepest_target)
  File "/full/path/WCT-TF2.3/vgg_normalised.py", line 35, in vgg_from_t7
    x = Conv2D(filters, kernel_size, padding='valid', activation=None, name=name,
  File "/home/user/.local/lib/python3.8/site-packages/tensorflow/python/keras/engine/base_layer.py", line 925, in __call__
    return self._functional_construction_call(inputs, args, kwargs,
  File "/home/user/.local/lib/python3.8/site-packages/tensorflow/python/keras/engine/base_layer.py", line 1098, in _functional_construction_call
    self._maybe_build(inputs)
  File "/home/user/.local/lib/python3.8/site-packages/tensorflow/python/keras/engine/base_layer.py", line 2643, in _maybe_build
    self.build(input_shapes)  # pylint:disable=not-callable
  File "/home/user/.local/lib/python3.8/site-packages/tensorflow/python/keras/layers/convolutional.py", line 197, in build
    self.kernel = self.add_weight(
  File "/home/user/.local/lib/python3.8/site-packages/tensorflow/python/keras/engine/base_layer.py", line 597, in add_weight
    variable = self._add_variable_with_custom_getter(
  File "/home/user/.local/lib/python3.8/site-packages/tensorflow/python/training/tracking/base.py", line 745, in _add_variable_with_custom_getter
    new_variable = getter(
  File "/home/user/.local/lib/python3.8/site-packages/tensorflow/python/keras/engine/base_layer_utils.py", line 133, in make_variable
    return tf_variables.VariableV1(
  File "/home/user/.local/lib/python3.8/site-packages/tensorflow/python/ops/variables.py", line 260, in __call__
    return cls._variable_v1_call(*args, **kwargs)
  File "/home/user/.local/lib/python3.8/site-packages/tensorflow/python/ops/variables.py", line 206, in _variable_v1_call
    return previous_getter(
  File "/home/user/.local/lib/python3.8/site-packages/tensorflow/python/ops/variables.py", line 199, in <lambda>
    previous_getter = lambda **kwargs: default_variable_creator(None, **kwargs)
  File "/home/user/.local/lib/python3.8/site-packages/tensorflow/python/ops/variable_scope.py", line 2583, in default_variable_creator
    return resource_variable_ops.ResourceVariable(
  File "/home/user/.local/lib/python3.8/site-packages/tensorflow/python/ops/variables.py", line 264, in __call__
    return super(VariableMetaclass, cls).__call__(*args, **kwargs)
  File "/home/user/.local/lib/python3.8/site-packages/tensorflow/python/ops/resource_variable_ops.py", line 1507, in __init__
    self._init_from_args(
  File "/home/user/.local/lib/python3.8/site-packages/tensorflow/python/ops/resource_variable_ops.py", line 1651, in _init_from_args
    initial_value() if init_from_fn else initial_value,
TypeError: <lambda>() got an unexpected keyword argument 'dtype'

Actually this log is after I converted the tree with tf_upgrade_v2. But the log after my manual editing was the same.
A command I used for running stylize.py:

python3 stylize.py --checkpoints models/relu5_1 models/relu4_1 models/relu3_1 models/relu2_1 models/relu1_1 --relu-targets relu5_1 relu4_1 relu3_1 relu2_1 relu1_1 --style-size 512 --alpha 0.8 --content-path imags --style-path styles --out-path output

An image is in imags directory, style image is in styles directory and output directory is empty one.

[aekasitt]

Had the same problem and found out that my Keras version was 2.3.1 not the recommended 2.0.X
So I fixed it by running

conda install keras=2.0.8

PS. This also means you need to be running Python 3.6

[thima2017]

Thank you. I would try it, but I don't know how to install python 3.6 on Ubuntu 20.04. And I got worked pytorch version (PytorchWCT) with python2 if someone have the same issue. I installed pip2 by an official script, old version of torch with suitable torchvision and it worked.