A Running Error Using Tensorflow 1.8.0

[LiChenyang-Github]

My tensorflow version is 1.8.0
I clone the code to my computer and download the mnist_cluttered_60x60_6distortions.npz dataset into ./data/.
When I run the main.py, I get the following ERROR:

Loading the data...
Building ConvNet...
Traceback (most recent call last):
  File "/home/xinlab/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/tensor_util.py", line 517, in make_tensor_proto
    str_values = [compat.as_bytes(x) for x in proto_values]
  File "/home/xinlab/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/tensor_util.py", line 517, in <listcomp>
    str_values = [compat.as_bytes(x) for x in proto_values]
  File "/home/xinlab/anaconda3/lib/python3.6/site-packages/tensorflow/python/util/compat.py", line 67, in as_bytes
    (bytes_or_text,))
TypeError: Expected binary or unicode string, got -1

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "main.py", line 422, in <module>
    main()
  File "main.py", line 275, in main
    pool3_flat, pool3_size = Flatten(pool3)
  File "/home/LiChenyang/spatial-transformer-network/utils/layer_utils.py", line 85, in Flatten
    layer_flat = tf.reshape(layer, [-1, num_features])
  File "/home/xinlab/anaconda3/lib/python3.6/site-packages/tensorflow/python/ops/gen_array_ops.py", line 6113, in reshape
    "Reshape", tensor=tensor, shape=shape, name=name)
  File "/home/xinlab/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py", line 513, in _apply_op_helper
    raise err
  File "/home/xinlab/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py", line 510, in _apply_op_helper
    preferred_dtype=default_dtype)
  File "/home/xinlab/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1104, in internal_convert_to_tensor
    ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
  File "/home/xinlab/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/constant_op.py", line 235, in _constant_tensor_conversion_function
    return constant(v, dtype=dtype, name=name)
  File "/home/xinlab/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/constant_op.py", line 214, in constant
    value, dtype=dtype, shape=shape, verify_shape=verify_shape))
  File "/home/xinlab/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/tensor_util.py", line 521, in make_tensor_proto
    "supported type." % (type(values), values))
TypeError: Failed to convert object of type <class 'list'> to Tensor. Contents: [-1, None]. Consider casting elements to a supported type.
LiChenyang@ubuntu:~/spatial-transformer-network$

Any idea to fix this problem?

[LiChenyang-Github]

I guess the error is due to the last dimension of the target shape is None(i.e. [-1, None]) when execute the code: layer_flat = tf.reshape(layer, [-1, num_features]).
I try to print the h_trans and pool3 in main(), and I get:

Tensor("AddN:0", shape=(?, ?, ?, 1), dtype=float32)
Tensor("Relu_5:0", shape=(?, ?, ?, 128), dtype=float32)

The second and third dimension of h_trans is '?', which makes num_features in Flatten() None.

To verify this idea, I add a new function in ./utils/layer_utils.py:

def Flatten_test(layer):
	"""
	Handy function for flattening the result of a conv2D or
	maxpool2D to be used for a fully-connected (affine) layer.
	"""
	# layer_shape = layer.get_shape()
	# # num_features = tf.reduce_prod(tf.shape(layer)[1:])
	# num_features = layer_shape[1:].num_elements()

	num_features = 8 * 8 * 128
	layer_flat = tf.reshape(layer, [-1, num_features])

	return layer_flat, num_features

Manually calculate the num_features and change:
pool3_flat, pool3_size = Flatten(pool3)
into
pool3_flat, pool3_size = Flatten_test(pool3)

Finally, I can successfully run the main.py.

However, I still puzzle about why the num_features is None in Flatten()