ranjanZ
commented
4 years ago Hello @ama542blake , Thank you for your interest. I can see that the in the tensor-flow version the parameter settings has been updated. See this https://www.tensorflow.org/api_docs/python/tf/nn/dilation2d. tf.nn.dilation2d( input, filters, strides, padding, data_format, dilations, name=None )
Hence you may update accordingly. For example: At line number 172 Change x = tf.nn.dilation2d(x, st_element, (1, ) + strides + (1, ), rates, padding.upper())
to
x = tf.nn.dilation2d(x, st_element, (1, ) + strides + (1, ), padding.upper(),"NHWC",rates)
Please check and let me know if there is any further problem in running.
ama542blake
Hello,
I have read your paper "Morphological Networks for Image De-raining”, and found it very interesting. I also have downloaded your code from the link to GitHub in your paper. Thank you very much for including that, it will be incredibly helpful. However I have set up the project and installed all of the requirements, but I am having trouble getting your code to run.
The first issue I have gives the following traceback: Traceback (most recent call last): File "run.py", line 3, in
from models import
File "/Users/Blake/Desktop/CV/Papers/Morphological Network/2D-Morphological-Network-master/src/models.py", line 8, in
from morph_layers2D import
File "/Users/Blake/Desktop/CV/Papers/Morphological Network/2D-Morphological-Network-master/src/morph_layers2D.py", line 5, in
from keras.layers import initializers,constraints
ImportError: cannot import name initializers
In doing some quick searches, I found I could get rid of the error by changing line 5 of morph_layers2D from
from keras.layers import initializers, constraintstofrom keras import initializers, constraintsBut after doing this, I get a new error
Traceback (most recent call last): File "run.py", line 163, in
main(input_dir=sys.argv[1],output_dir=sys.argv[2])
File "run.py", line 128, in main
D=get_all_models()
File "run.py", line 56, in get_all_models
path1=model_path1()
File "/Users/Blake/Desktop/CV/Papers/Morphological Network/2D-Morphological-Network-master/src/models.py", line 86, in model_path1
z1=Dilation2D(4, (8,8),padding="same",strides=(1,1))(z1)
File "/Users/Blake/anaconda3/envs/py2/lib/python2.7/site-packages/keras/backend/tensorflow_backend.py", line 75, in symbolic_fn_wrapper
return func(*args, kwargs)
File "/Users/Blake/anaconda3/envs/py2/lib/python2.7/site-packages/keras/engine/base_layer.py", line 489, in call
output = self.call(inputs, kwargs)
File "/Users/Blake/Desktop/CV/Papers/Morphological Network/2D-Morphological-Network-master/src/morph_layers2D.py", line 143, in call
self.strides, self.padding),
File "/Users/Blake/Desktop/CV/Papers/Morphological Network/2D-Morphological-Network-master/src/morph_layers2D.py", line 172, in __dilation2d
rates, padding.upper())
TypeError: dilation2d_v2() takes at least 6 arguments (5 given)
Running on Python 2.7.16, Keras 2.3.1, tesorflow 2.0.0
Going through all of the locations mentioned in the traceback, I can not find anything with the name dilation2d_v2. How can I solve this issue?