ftokarev / tf-adain Goto Github PK

Hi,

Could you please provide an example on how to use a trained model on our own dataset?

I am taking the following steps:

1. After the checkpoint is created, I pass the checkpoint to test.py as "decoder_weights".
2. I had to change "saver.restore(sess, decoder_weights)" to "saver.restore(sess, tf.train.latest_checkpoint(decoder_weights))" to read it successfully from checkpoint directory.
3. What should be passed to vgg_weights in test.py? Are we not fine-tuning the encoder (i.e. vgg weights) in the training on our own dataset?
   Does it remain same as "models/vgg19_weights_normalized.h5"?

Thank you in advance.

Could you provide the last checkpoint please to resume it in training process ?

19/5000
I have encountered this error when I run test.py with cpu:
ValueError: Cannot feed value of shape (2, 3, 380, 600) for Tensor 'Placeholder:0', which has shape '(?, ?, ?, 3)'

Hi,

In test.py in CPU mode the call to prepare_image line (178 and 179) returns wrong image shape due to wrong data_format.
To fix just pass the data_format para:

style_image = prepare_image(style_image, True, data_format)
content_image = prepare_image(content_image, True, data_format)

Traceback (most recent call last):
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py", line 1356, in _do_call
return fn(*args)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py", line 1341, in _run_fn
options, feed_dict, fetch_list, target_list, run_metadata)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py", line 1429, in _call_tf_sessionrun
run_metadata)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Default MaxPoolingOp only supports NHWC on device type CPU
[[{{node vgg/pool1/MaxPool}}]]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "test.py", line 274, in
style_transfer(**vars(args))
File "test.py", line 181, in style_transfer
image: style_image[np.newaxis,:]
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py", line 950, in run
run_metadata_ptr)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py", line 1173, in _run
feed_dict_tensor, options, run_metadata)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py", line 1350, in _do_run
run_metadata)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py", line 1370, in _do_call
raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Default MaxPoolingOp only supports NHWC on device type CPU
[[node vgg/pool1/MaxPool (defined at /tmp/tmpsb0fdgud.py:30) ]]

Errors may have originated from an input operation.
Input Source operations connected to node vgg/pool1/MaxPool:
vgg/conv1_2/Relu (defined at /tmp/tmp649k6325.py:74)

Original stack trace for 'vgg/pool1/MaxPool':
File "test.py", line 274, in
style_transfer(**vars(args))
File "test.py", line 83, in style_transfer
decoder_weights if decoder_in_h5 else None, alpha, data_format=data_format)
File "test.py", line 216, in build_graph
vgg = build_vgg(image, w, data_format=data_format)
File "/home/nd/workspace/workspace_lsh/tf-adain/adain/nn.py", line 54, in build_vgg
activation=tf.nn.relu, trainable=False, data_format=data_format)
File "/home/nd/workspace/workspace_lsh/tf-adain/adain/nn.py", line 102, in build_net
data_format=data_format)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/util/deprecation.py", line 324, in new_func
return func(*args, **kwargs)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/layers/pooling.py", line 311, in max_pooling2d
return layer.apply(inputs)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 1479, in apply
return self.call(inputs, *args, **kwargs)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/layers/base.py", line 537, in call
outputs = super(Layer, self).call(inputs, *args, **kwargs)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/keras/engine/base_layer.py", line 634, in call
outputs = call_fn(inputs, *args, **kwargs)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/autograph/impl/api.py", line 146, in wrapper
), args, kwargs)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/autograph/impl/api.py", line 450, in converted_call
result = converted_f(*effective_args, **kwargs)
File "/tmp/tmpsb0fdgud.py", line 30, in tf__call
outputs = ag.converted_call('pool_function', self, ag__.ConversionOptions(recursive=True, force_conversion=False, optional_features=(), internal_convert_user_code=True), (inputs,), {'ksize': pool_shape, 'strides': strides, 'padding': ag__.converted_call('upper', self.padding, ag__.ConversionOptions(recursive=True, force_conversion=False, optional_features=(), internal_convert_user_code=True), (), None), 'data_format': ag__.converted_call('convert_data_format', conv_utils, ag__.ConversionOptions(recursive=True, force_conversion=False, optional_features=(), internal_convert_user_code=True), (self.data_format, 4), None)})
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/autograph/impl/api.py", line 356, in converted_call
return _call_unconverted(f, args, kwargs)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/autograph/impl/api.py", line 253, in _call_unconverted
return f(*args, **kwargs)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/nn_ops.py", line 3756, in max_pool
name=name)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/gen_nn_ops.py", line 5672, in max_pool
data_format=data_format, name=name)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/op_def_library.py", line 788, in _apply_op_helper
op_def=op_def)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/util/deprecation.py", line 507, in new_func
return func(*args, **kwargs)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/ops.py", line 3616, in create_op
op_def=op_def)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/ops.py", line 2005, in init
self._traceback = tf_stack.extract_stack()