iwyoo/dirnet-tensorflow

Tensorflow implementation of DIRNet

dirnet-tensorflow's Issues

Getting error after executing command "python train.py"

Hi

Please look at the following error which I got when I tried to train the network.

File "train.py", line 26, in
main()
File "train.py", line 14, in main
dh = MNISTDataHandler("MNIST_data", is_train=True)
File "/home/uavws/Desktop/kanti/NN/tesnsorflow/DIRNet/data.py", line 13, in init
self.data = self._get_data()
File "/home/uavws/Desktop/kanti/NN/tesnsorflow/DIRNet/data.py", line 31, in _get_data
images = extract_images(f)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/contrib/learn/python/learn/datasets/mnist.py", line 55, in extract_images
magic = _read32(bytestream)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/contrib/learn/python/learn/datasets/mnist.py", line 37, in _read32
return numpy.frombuffer(bytestream.read(4), dtype=dt)[0]
File "/usr/lib/python2.7/gzip.py", line 261, in read
self._read(readsize)
File "/usr/lib/python2.7/gzip.py", line 296, in _read
self._read_gzip_header()
File "/usr/lib/python2.7/gzip.py", line 190, in _read_gzip_header
raise IOError, 'Not a gzipped file'
IOError: Not a gzipped file

Looking forward to hear from you soon.

questions

I have 3 questions:

1. why do you choose 10,000 iterations? I think this is big number for small images and it takes about 2 days to execute without GPU.
2. How can we calculate the algorithm accuracy?
3. How can we modify the code to find transformation parameters between fixed and floating images in case of image registration?

Thanks for sharing your code

Error of annotation !

DIRNet-tensorflow/WarpST.py

Line 123 in 0f8b403

# transform (x, y)^T -> (x+vx, x+vy)^T

this line is incorrect, the coordinate of the regular grid (x, y) are transformed into (x+v_x, y+v_y), where (v_x, v_y) are the displacement vector of the deformable field that is interpolated by bicubic interpolation. Is there any mistake in my understanding?

How to evaluate the algorithm by the average 2D Euclidean distance between a landmark in the warped image and its corresponding landmark in the fixed image?

If I feed the the pair of the test images(fixed image and moved image) into the model trained well. I can easily get the registed image. But, If I select several point in the moved image, their coordinates like that (x1, y1), (x2, y2) ... (x10, y10). Thoes points put into the model trained well and do this deformable transformation. After that, I want to know how I can get thoes changed coordinates. I want to evaluate the algorithm by calculating the average 2D Euclidean distance between a landmark in the warped image and its corresponding landmark in the fixed image . Thanks in advance.

Troubleshoot Error with Tensorflow-Python

Hi,

I am getting following error message when executing a python project in Jupiter using Tensorflow version 1.15.2 on Mac OS. Please suggest me any solution for this error.

"ValueError: Tried to convert 'input' to a tensor and failed. Error: None values not supported."

ValueError Traceback (most recent call last)

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/op_def_library.py in _apply_op_helper(self, op_type_name, name, **keywords)

527                 as_ref=input_arg.is_ref,

--> 528 preferred_dtype=default_dtype)

529           except TypeError as err:

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/ops.py in internal_convert_to_tensor(value, dtype, name, as_ref, preferred_dtype, ctx, accepted_result_types)

1296 if ret is None:

-> 1297 ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)

1298

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_tensor_conversion_function(v, dtype, name, as_ref)

285   _ = as_ref

--> 286 return constant(v, dtype=dtype, name=name)

287

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in constant(value, dtype, shape, name)

226   return _constant_impl(value, dtype, shape, name, verify_shape=False,

--> 227 allow_broadcast=True)

228

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_impl(value, dtype, shape, name, verify_shape, allow_broadcast)

264           value, dtype=dtype, shape=shape, verify_shape=verify_shape,

--> 265 allow_broadcast=allow_broadcast))

266   dtype_value = attr_value_pb2.AttrValue(type=tensor_value.tensor.dtype)

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/tensor_util.py in make_tensor_proto(values, dtype, shape, verify_shape, allow_broadcast)

436     if values is None:

--> 437 raise ValueError("None values not supported.")

438     # if dtype is provided, forces numpy array to be the type

ValueError: None values not supported.

During handling of the above exception, another exception occurred:

ValueError Traceback (most recent call last)

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/op_def_library.py in _apply_op_helper(self, op_type_name, name, **keywords)

541               observed = ops.internal_convert_to_tensor(

--> 542 values, as_ref=input_arg.is_ref).dtype.name

543             except ValueError as err:

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/ops.py in internal_convert_to_tensor(value, dtype, name, as_ref, preferred_dtype, ctx, accepted_result_types)

1296 if ret is None:

-> 1297 ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)

1298

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_tensor_conversion_function(v, dtype, name, as_ref)

285   _ = as_ref

--> 286 return constant(v, dtype=dtype, name=name)

287

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in constant(value, dtype, shape, name)

226   return _constant_impl(value, dtype, shape, name, verify_shape=False,

--> 227 allow_broadcast=True)

228

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_impl(value, dtype, shape, name, verify_shape, allow_broadcast)

264           value, dtype=dtype, shape=shape, verify_shape=verify_shape,

--> 265 allow_broadcast=allow_broadcast))

266   dtype_value = attr_value_pb2.AttrValue(type=tensor_value.tensor.dtype)

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/tensor_util.py in make_tensor_proto(values, dtype, shape, verify_shape, allow_broadcast)

436     if values is None:

--> 437 raise ValueError("None values not supported.")

438     # if dtype is provided, forces numpy array to be the type

ValueError: None values not supported.

During handling of the above exception, another exception occurred:

ValueError Traceback (most recent call last)

in

 24

 25 if __name__ == "__main__":

---> 26 main()

in main()

 11   mkdir(config.ckpt_dir)

 12

---> 13 reg = DIRNet(sess, config, "DIRNet", is_train=True)

 14   dh = MNISTDataHandler("MNIST_data", is_train=True)

 15

~/Documents/DIRNet-tensorflow 4.5.1/models.py in init(self, sess, config, name, is_train)

 63

 64     if self.is_train :

---> 65 self.loss = ncc(self.y, self.z)

 66       #self.loss = mse(self.y, self.z)

 67

~/Documents/DIRNet-tensorflow 4.5.1/ops.py in ncc(x, y)

 33 def ncc(x, y):

 34   mean_x = tf.reduce_mean(x, [1,2,3], keep_dims=True)

---> 35 mean_y = tf.reduce_mean(y, [1,2,3], keep_dims=True)

 36   mean_x2 = tf.reduce_mean(tf.square(x), [1,2,3], keep_dims=True)

 37   mean_y2 = tf.reduce_mean(tf.square(y), [1,2,3], keep_dims=True)

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/ops/math_ops.py in reduce_mean_v1(input_tensor, axis, keepdims, name, reduction_indices, keep_dims)

1810 keepdims = deprecation.deprecated_argument_lookup("keepdims", keepdims,

1811 "keep_dims", keep_dims)

-> 1812 return reduce_mean(input_tensor, axis, keepdims, name)

1813

1814

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/util/dispatch.py in wrapper(*args, **kwargs)

178     """Call target, and fall back on dispatchers if there is a TypeError."""

179     try:

--> 180 return target(*args, **kwargs)

181     except (TypeError, ValueError):

182       # Note: convert_to_eager_tensor currently raises a ValueError, not a

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/ops/math_ops.py in reduce_mean(input_tensor, axis, keepdims, name)

1868 gen_math_ops.mean(

1869 input_tensor, _ReductionDims(input_tensor, axis), keepdims,

-> 1870 name=name))

1871

1872

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/ops/gen_math_ops.py in mean(input, axis, keep_dims, name)

6389 _, _, _op = _op_def_lib._apply_op_helper(

6390 "Mean", input=input, reduction_indices=axis, keep_dims=keep_dims,

-> 6391 name=name)

6392 _result = _op.outputs[:]

6393 _inputs_flat = _op.inputs

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/op_def_library.py in _apply_op_helper(self, op_type_name, name, **keywords)

544               raise ValueError(

545                   "Tried to convert '%s' to a tensor and failed. Error: %s" %

--> 546 (input_name, err))

547             prefix = ("Input '%s' of '%s' Op has type %s that does not match" %

548                       (input_name, op_type_name, observed))

ValueError: Tried to convert 'input' to a tensor and failed. Error: None values not supported.

I need some help for understanding the wrapST

Hello,I have some questions about the code.I'd appreciate that if you have time to help me .

self.v = self.vCNN(self.xy)
self.z = WarpST(self.x, self.v, config.im_size)

the self.v is a (64,7,7,2)tensor,I really could't understand the following coda:

# grid of (x_t, y_t, 1), eq (1) in ref [1]
            height_f = tf.cast(height, 'float32')
            width_f = tf.cast(width, 'float32')
            out_height = out_size[0]
            out_width = out_size[1]
            grid = _meshgrid(out_height, out_width)     # [2, h*w]      [2,784]
            grid = tf.reshape(grid, [-1])               # [2*h*w]       [1568,]
            grid = tf.tile(grid, tf.stack([num_batch]))           # [n*2*h*w]
            grid = tf.reshape(grid, tf.stack([num_batch, 2, -1])) # [n, 2, h*w]     [?,2,?]

            # Set source position (x+vx, y+vy)^T
            V = bicubic_interp_2d(V, out_size)                          # [64,28,28,2]
            V = tf.transpose(V, [0, 3, 1, 2])           # [n, 2, h, w]     [64,2,28,28]
            V = tf.reshape(V, [num_batch, 2, -1])       # [n, 2, h*w]
            T_g = tf.add(V, grid)                       # [n, 2, h*w]

            x_s = tf.slice(T_g, [0, 0, 0], [-1, 1, -1])
            y_s = tf.slice(T_g, [0, 1, 0], [-1, 1, -1])    
            x_s_flat = tf.reshape(x_s, [-1])        # [?,]
            y_s_flat = tf.reshape(y_s, [-1])        # [?,]

            input_transformed = _interpolate(
                U, x_s_flat, y_s_flat, out_size)

What is the function of the T_g and x_s_flat？

ValueError: Tried to convert 'input' to a tensor and failed. Error: None values not supported

I am researching on a project Image Registration on Brain Haemorrhage but encountered an Error : ValueError: Tried to convert 'input' to a tensor and failed. Error: None values not supported. .I am using Tensorflow 1.15.2. Recommend some means to troubleShoot The issue.

WARNING:tensorflow:From /Users/subha/Documents/DIRNet-tensorflow 4.5.1/models.py:48: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.

WARNING:tensorflow:From /Users/subha/Documents/DIRNet-tensorflow 4.5.1/models.py:13: The name tf.variable_scope is deprecated. Please use tf.compat.v1.variable_scope instead.

WARNING:tensorflow:From /Users/subha/Documents/DIRNet-tensorflow 4.5.1/ops.py:8: The name tf.get_variable is deprecated. Please use tf.compat.v1.get_variable instead.

WARNING:tensorflow:
The TensorFlow contrib module will not be included in TensorFlow 2.0.
For more information, please see:

• https://github.com/tensorflow/community/blob/master/rfcs/20180907-contrib-sunset.md
• https://github.com/tensorflow/addons
• https://github.com/tensorflow/io (for I/O related ops)
  If you depend on functionality not listed there, please file an issue.

WARNING:tensorflow:From /Users/subha/Documents/DIRNet-tensorflow 4.5.1/models.py:16: The name tf.nn.avg_pool is deprecated. Please use tf.nn.avg_pool2d instead.

WARNING:tensorflow:From /Users/subha/Documents/DIRNet-tensorflow 4.5.1/models.py:27: The name tf.get_collection is deprecated. Please use tf.compat.v1.get_collection instead.

WARNING:tensorflow:From /Users/subha/Documents/DIRNet-tensorflow 4.5.1/models.py:28: The name tf.GraphKeys is deprecated. Please use tf.compat.v1.GraphKeys instead.

WARNING:tensorflow:From /Users/subha/Documents/DIRNet-tensorflow 4.5.1/models.py:29: The name tf.train.Saver is deprecated. Please use tf.compat.v1.train.Saver instead.

---

ValueError Traceback (most recent call last)
/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/op_def_library.py in _apply_op_helper(self, op_type_name, name, **keywords)
527 as_ref=input_arg.is_ref,
--> 528 preferred_dtype=default_dtype)
529 except TypeError as err:

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/ops.py in internal_convert_to_tensor(value, dtype, name, as_ref, preferred_dtype, ctx, accepted_result_types)
1296 if ret is None:
-> 1297 ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
1298

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_tensor_conversion_function(v, dtype, name, as_ref)
285 _ = as_ref
--> 286 return constant(v, dtype=dtype, name=name)
287

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in constant(value, dtype, shape, name)
226 return _constant_impl(value, dtype, shape, name, verify_shape=False,
--> 227 allow_broadcast=True)
228

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_impl(value, dtype, shape, name, verify_shape, allow_broadcast)
264 value, dtype=dtype, shape=shape, verify_shape=verify_shape,
--> 265 allow_broadcast=allow_broadcast))
266 dtype_value = attr_value_pb2.AttrValue(type=tensor_value.tensor.dtype)

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/tensor_util.py in make_tensor_proto(values, dtype, shape, verify_shape, allow_broadcast)
436 if values is None:
--> 437 raise ValueError("None values not supported.")
438 # if dtype is provided, forces numpy array to be the type

ValueError: None values not supported.

During handling of the above exception, another exception occurred:

ValueError Traceback (most recent call last)
/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/op_def_library.py in _apply_op_helper(self, op_type_name, name, **keywords)
541 observed = ops.internal_convert_to_tensor(
--> 542 values, as_ref=input_arg.is_ref).dtype.name
543 except ValueError as err:

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/ops.py in internal_convert_to_tensor(value, dtype, name, as_ref, preferred_dtype, ctx, accepted_result_types)
1296 if ret is None:
-> 1297 ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
1298

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_tensor_conversion_function(v, dtype, name, as_ref)
285 _ = as_ref
--> 286 return constant(v, dtype=dtype, name=name)
287

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in constant(value, dtype, shape, name)
226 return _constant_impl(value, dtype, shape, name, verify_shape=False,
--> 227 allow_broadcast=True)
228

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_impl(value, dtype, shape, name, verify_shape, allow_broadcast)
264 value, dtype=dtype, shape=shape, verify_shape=verify_shape,
--> 265 allow_broadcast=allow_broadcast))
266 dtype_value = attr_value_pb2.AttrValue(type=tensor_value.tensor.dtype)

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/tensor_util.py in make_tensor_proto(values, dtype, shape, verify_shape, allow_broadcast)
436 if values is None:
--> 437 raise ValueError("None values not supported.")
438 # if dtype is provided, forces numpy array to be the type

ValueError: None values not supported.

During handling of the above exception, another exception occurred:

ValueError Traceback (most recent call last)
in
24
25 if name == "main":
---> 26 main()

in main()
11 mkdir(config.ckpt_dir)
12
---> 13 reg = DIRNet(sess, config, "DIRNet", is_train=True)
14 dh = MNISTDataHandler("MNIST_data", is_train=True)
15

~/Documents/DIRNet-tensorflow 4.5.1/models.py in init(self, sess, config, name, is_train)
63
64 if self.is_train :
---> 65 self.loss = ncc(self.y, self.z)
66 #self.loss = mse(self.y, self.z)
67

~/Documents/DIRNet-tensorflow 4.5.1/ops.py in ncc(x, y)
33 def ncc(x, y):
34 mean_x = tf.reduce_mean(x, [1,2,3], keep_dims=True)
---> 35 mean_y = tf.reduce_mean(y, [1,2,3], keep_dims=True)
36 mean_x2 = tf.reduce_mean(tf.square(x), [1,2,3], keep_dims=True)
37 mean_y2 = tf.reduce_mean(tf.square(y), [1,2,3], keep_dims=True)

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/ops/math_ops.py in reduce_mean_v1(input_tensor, axis, keepdims, name, reduction_indices, keep_dims)
1810 keepdims = deprecation.deprecated_argument_lookup("keepdims", keepdims,
1811 "keep_dims", keep_dims)
-> 1812 return reduce_mean(input_tensor, axis, keepdims, name)
1813
1814

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/util/dispatch.py in wrapper(*args, **kwargs)
178 """Call target, and fall back on dispatchers if there is a TypeError."""
179 try:
--> 180 return target(*args, **kwargs)
181 except (TypeError, ValueError):
182 # Note: convert_to_eager_tensor currently raises a ValueError, not a

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/ops/math_ops.py in reduce_mean(input_tensor, axis, keepdims, name)
1868 gen_math_ops.mean(
1869 input_tensor, _ReductionDims(input_tensor, axis), keepdims,
-> 1870 name=name))
1871
1872

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/ops/gen_math_ops.py in mean(input, axis, keep_dims, name)
6389 _, _, _op = _op_def_lib._apply_op_helper(
6390 "Mean", input=input, reduction_indices=axis, keep_dims=keep_dims,
-> 6391 name=name)
6392 _result = _op.outputs[:]
6393 _inputs_flat = _op.inputs

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/op_def_library.py in _apply_op_helper(self, op_type_name, name, **keywords)
544 raise ValueError(
545 "Tried to convert '%s' to a tensor and failed. Error: %s" %
--> 546 (input_name, err))
547 prefix = ("Input '%s' of '%s' Op has type %s that does not match" %
548 (input_name, op_type_name, observed))

ValueError: Tried to convert 'input' to a tensor and failed. Error: None values not supported.

ValueError: Tried to convert 'input' to a tensor and failed. Error: None values not supported.

I have tried to apply DIRNet on Brain Haemorrhage Datasets but encountered an error while training .I could resolve the Error: ValueError: Tried to convert 'input' to a tensor and failed. Error: None values not supported.' .Need your recommendation if any .I am using Tensorflow 1.15.2.

ValueError Traceback (most recent call last)
/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/op_def_library.py in _apply_op_helper(self, op_type_name, name, **keywords)
527 as_ref=input_arg.is_ref,
--> 528 preferred_dtype=default_dtype)
529 except TypeError as err:

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/ops.py in internal_convert_to_tensor(value, dtype, name, as_ref, preferred_dtype, ctx, accepted_result_types)
1296 if ret is None:
-> 1297 ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
1298

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_tensor_conversion_function(v, dtype, name, as_ref)
285 _ = as_ref
--> 286 return constant(v, dtype=dtype, name=name)
287

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in constant(value, dtype, shape, name)
226 return _constant_impl(value, dtype, shape, name, verify_shape=False,
--> 227 allow_broadcast=True)
228

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_impl(value, dtype, shape, name, verify_shape, allow_broadcast)
264 value, dtype=dtype, shape=shape, verify_shape=verify_shape,
--> 265 allow_broadcast=allow_broadcast))
266 dtype_value = attr_value_pb2.AttrValue(type=tensor_value.tensor.dtype)

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/tensor_util.py in make_tensor_proto(values, dtype, shape, verify_shape, allow_broadcast)
436 if values is None:
--> 437 raise ValueError("None values not supported.")
438 # if dtype is provided, forces numpy array to be the type

ValueError: None values not supported.

During handling of the above exception, another exception occurred:

ValueError Traceback (most recent call last)
/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/op_def_library.py in _apply_op_helper(self, op_type_name, name, **keywords)
541 observed = ops.internal_convert_to_tensor(
--> 542 values, as_ref=input_arg.is_ref).dtype.name
543 except ValueError as err:

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/ops.py in internal_convert_to_tensor(value, dtype, name, as_ref, preferred_dtype, ctx, accepted_result_types)
1296 if ret is None:
-> 1297 ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
1298

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_tensor_conversion_function(v, dtype, name, as_ref)
285 _ = as_ref
--> 286 return constant(v, dtype=dtype, name=name)
287

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in constant(value, dtype, shape, name)
226 return _constant_impl(value, dtype, shape, name, verify_shape=False,
--> 227 allow_broadcast=True)
228

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_impl(value, dtype, shape, name, verify_shape, allow_broadcast)
264 value, dtype=dtype, shape=shape, verify_shape=verify_shape,
--> 265 allow_broadcast=allow_broadcast))
266 dtype_value = attr_value_pb2.AttrValue(type=tensor_value.tensor.dtype)

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/tensor_util.py in make_tensor_proto(values, dtype, shape, verify_shape, allow_broadcast)
436 if values is None:
--> 437 raise ValueError("None values not supported.")
438 # if dtype is provided, forces numpy array to be the type

ValueError: None values not supported.

During handling of the above exception, another exception occurred:

ValueError Traceback (most recent call last)
in
24
25 if name == "main":
---> 26 main()

in main()
11 mkdir(config.ckpt_dir)
12
---> 13 reg = DIRNet(sess, config, "DIRNet", is_train=True)
14 dh = MNISTDataHandler("MNIST_data", is_train=True)
15

~/Documents/DIRNet-tensorflow 4.5.1/models.py in init(self, sess, config, name, is_train)
64
65 if self.is_train :
---> 66 self.loss = ncc(self.y, self.z)
67 #self.loss = mse(self.y, self.z)
68

~/Documents/DIRNet-tensorflow 4.5.1/ops.py in ncc(x, y)
33 def ncc(x, y):
34 mean_x = tf.reduce_mean(x, [1,2,3], keep_dims=True)
---> 35 mean_y = tf.reduce_mean(y, [1,2,3], keep_dims=True)
36 mean_x2 = tf.reduce_mean(tf.square(x), [1,2,3], keep_dims=True)
37 mean_y2 = tf.reduce_mean(tf.square(y), [1,2,3], keep_dims=True)

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/ops/math_ops.py in reduce_mean_v1(input_tensor, axis, keepdims, name, reduction_indices, keep_dims)
1810 keepdims = deprecation.deprecated_argument_lookup("keepdims", keepdims,
1811 "keep_dims", keep_dims)
-> 1812 return reduce_mean(input_tensor, axis, keepdims, name)
1813
1814

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/util/dispatch.py in wrapper(*args, **kwargs)
178 """Call target, and fall back on dispatchers if there is a TypeError."""
179 try:
--> 180 return target(*args, **kwargs)
181 except (TypeError, ValueError):
182 # Note: convert_to_eager_tensor currently raises a ValueError, not a

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/ops/math_ops.py in reduce_mean(input_tensor, axis, keepdims, name)
1868 gen_math_ops.mean(
1869 input_tensor, _ReductionDims(input_tensor, axis), keepdims,
-> 1870 name=name))
1871
1872

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/ops/gen_math_ops.py in mean(input, axis, keep_dims, name)
6389 _, _, _op = _op_def_lib._apply_op_helper(
6390 "Mean", input=input, reduction_indices=axis, keep_dims=keep_dims,
-> 6391 name=name)
6392 _result = _op.outputs[:]
6393 _inputs_flat = _op.inputs

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/op_def_library.py in _apply_op_helper(self, op_type_name, name, **keywords)
544 raise ValueError(
545 "Tried to convert '%s' to a tensor and failed. Error: %s" %
--> 546 (input_name, err))
547 prefix = ("Input '%s' of '%s' Op has type %s that does not match" %
548 (input_name, op_type_name, observed))

ValueError: Tried to convert 'input' to a tensor and failed. Error: None values not supported.

How to visualize the displacement vector field?

Given a pair of images, the convolutional neural network outputs a displacement vector field(DVF). The DVF is a numpy array with two channels. If I want to visualize it, what should I do ？Opencv and matplotlib can not do this . Can you help me? Thanks very much.

How are the coordinates of the points in the image transformed through deformable transformation？　

I do not know how the coordinates of the points in the moved image are changed after deformable transformation. Could you give me some help? If I want to get the coordinates of the points after deformable transformation, how to implement it ?

Tensorflow Version

If we apply AffineST What is Version of TF.

Which transformation is used here?

Hi,
Just a short question: Which transformation is implemented in this DIRnet? Is it a cubic B spline transformation as described in "Nonrigid Registration Using Free-Form
Deformations: Application to Breast MR Images" (Section II B) and mentioned in the original DIRnet paper?

Thanks in advance, and nice code!
Cheers

Custom Dataset

Hi @iwyoo,

thank you for this repository!

Is there a way to train the model with a custom set of images?
If yes, how can I do that?

Thanks.

What is the maximum input image size?

Hello contributors.

I tried it.
It seems that it works well with the same 28 x 28 image as MNIST.
However, black pixels are displayed in the periphery.
After that, I tried to input an image of about 256 x 256, but the image was failing.
Since this network had only 4 convolution layers, I thought it was not dependent on image size, but that did not work.
Is this the limit of the model? I do not know whether it is a problem of post processing.
Is it possible to input large images?

import tensorflow as tf
from models import DIRNet
from config import get_config
from ops import mkdir
import cv2
import numpy as np

def main():
    sess = tf.Session()
    img_gray_x = np.array(cv2.imread('frame_000000.png', cv2.IMREAD_GRAYSCALE)) / 255.0
    img_gray_y = np.array(cv2.imread('frame_000001.png', cv2.IMREAD_GRAYSCALE)) / 255.0
    batch_x = img_gray_x[np.newaxis, 500:756, 1000:1256,np.newaxis]
    batch_y = img_gray_y[np.newaxis, 500:756, 1000:1256,np.newaxis]

    config = get_config(is_train=False)
    config.im_size = [256, 256]
    config.batch_size = 1

    mkdir(config.result_dir)

    reg = DIRNet(sess, config, "DIRNet", is_train=False)
    reg.restore(config.ckpt_dir)

    result_i_dir = config.result_dir+"/img"
    mkdir(result_i_dir)

    reg.deploy(result_i_dir, batch_x, batch_y)

if __name__ == "__main__":
    main()

What version of tensorflow is used?

ValueError: Tried to convert 'input' to a tensor and failed. Error: None values not supported.

I have tried to apply DIRNet on Brain Haemorrhage Datasets but encountered an error while training .I could resolve the Error: ValueError: Tried to convert 'input' to a tensor and failed. Error: None values not supported.' .Need your recommendation if any .I am using Tensorflow 1.15.2.

ValueError Traceback (most recent call last)
/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/op_def_library.py in _apply_op_helper(self, op_type_name, name, **keywords)
527 as_ref=input_arg.is_ref,
--> 528 preferred_dtype=default_dtype)
529 except TypeError as err:

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/ops.py in internal_convert_to_tensor(value, dtype, name, as_ref, preferred_dtype, ctx, accepted_result_types)
1296 if ret is None:
-> 1297 ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
1298

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_tensor_conversion_function(v, dtype, name, as_ref)
285 _ = as_ref
--> 286 return constant(v, dtype=dtype, name=name)
287

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in constant(value, dtype, shape, name)
226 return _constant_impl(value, dtype, shape, name, verify_shape=False,
--> 227 allow_broadcast=True)
228

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_impl(value, dtype, shape, name, verify_shape, allow_broadcast)
264 value, dtype=dtype, shape=shape, verify_shape=verify_shape,
--> 265 allow_broadcast=allow_broadcast))
266 dtype_value = attr_value_pb2.AttrValue(type=tensor_value.tensor.dtype)

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/tensor_util.py in make_tensor_proto(values, dtype, shape, verify_shape, allow_broadcast)
436 if values is None:
--> 437 raise ValueError("None values not supported.")
438 # if dtype is provided, forces numpy array to be the type

ValueError: None values not supported.

During handling of the above exception, another exception occurred:

ValueError Traceback (most recent call last)
/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/op_def_library.py in _apply_op_helper(self, op_type_name, name, **keywords)
541 observed = ops.internal_convert_to_tensor(
--> 542 values, as_ref=input_arg.is_ref).dtype.name
543 except ValueError as err:

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/ops.py in internal_convert_to_tensor(value, dtype, name, as_ref, preferred_dtype, ctx, accepted_result_types)
1296 if ret is None:
-> 1297 ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
1298

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_tensor_conversion_function(v, dtype, name, as_ref)
285 _ = as_ref
--> 286 return constant(v, dtype=dtype, name=name)
287

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in constant(value, dtype, shape, name)
226 return _constant_impl(value, dtype, shape, name, verify_shape=False,
--> 227 allow_broadcast=True)
228

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/constant_op.py in _constant_impl(value, dtype, shape, name, verify_shape, allow_broadcast)
264 value, dtype=dtype, shape=shape, verify_shape=verify_shape,
--> 265 allow_broadcast=allow_broadcast))
266 dtype_value = attr_value_pb2.AttrValue(type=tensor_value.tensor.dtype)

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/tensor_util.py in make_tensor_proto(values, dtype, shape, verify_shape, allow_broadcast)
436 if values is None:
--> 437 raise ValueError("None values not supported.")
438 # if dtype is provided, forces numpy array to be the type

ValueError: None values not supported.

During handling of the above exception, another exception occurred:

ValueError Traceback (most recent call last)
in
24
25 if name == "main":
---> 26 main()

in main()
11 mkdir(config.ckpt_dir)
12
---> 13 reg = DIRNet(sess, config, "DIRNet", is_train=True)
14 dh = MNISTDataHandler("MNIST_data", is_train=True)
15

~/Documents/DIRNet-tensorflow 4.5.1/models.py in init(self, sess, config, name, is_train)
64
65 if self.is_train :
---> 66 self.loss = ncc(self.y, self.z)
67 #self.loss = mse(self.y, self.z)
68

~/Documents/DIRNet-tensorflow 4.5.1/ops.py in ncc(x, y)
33 def ncc(x, y):
34 mean_x = tf.reduce_mean(x, [1,2,3], keep_dims=True)
---> 35 mean_y = tf.reduce_mean(y, [1,2,3], keep_dims=True)
36 mean_x2 = tf.reduce_mean(tf.square(x), [1,2,3], keep_dims=True)
37 mean_y2 = tf.reduce_mean(tf.square(y), [1,2,3], keep_dims=True)

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/ops/math_ops.py in reduce_mean_v1(input_tensor, axis, keepdims, name, reduction_indices, keep_dims)
1810 keepdims = deprecation.deprecated_argument_lookup("keepdims", keepdims,
1811 "keep_dims", keep_dims)
-> 1812 return reduce_mean(input_tensor, axis, keepdims, name)
1813
1814

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/util/dispatch.py in wrapper(*args, **kwargs)
178 """Call target, and fall back on dispatchers if there is a TypeError."""
179 try:
--> 180 return target(*args, **kwargs)
181 except (TypeError, ValueError):
182 # Note: convert_to_eager_tensor currently raises a ValueError, not a

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/ops/math_ops.py in reduce_mean(input_tensor, axis, keepdims, name)
1868 gen_math_ops.mean(
1869 input_tensor, _ReductionDims(input_tensor, axis), keepdims,
-> 1870 name=name))
1871
1872

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/ops/gen_math_ops.py in mean(input, axis, keep_dims, name)
6389 _, _, _op = _op_def_lib._apply_op_helper(
6390 "Mean", input=input, reduction_indices=axis, keep_dims=keep_dims,
-> 6391 name=name)
6392 _result = _op.outputs[:]
6393 _inputs_flat = _op.inputs

/anaconda3/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/op_def_library.py in _apply_op_helper(self, op_type_name, name, **keywords)
544 raise ValueError(
545 "Tried to convert '%s' to a tensor and failed. Error: %s" %
--> 546 (input_name, err))
547 prefix = ("Input '%s' of '%s' Op has type %s that does not match" %
548 (input_name, op_type_name, observed))

ValueError: Tried to convert 'input' to a tensor and failed. Error: None values not supported.

Wh