Operands could not be broadcast together with shapes (13, 128, 13) (13, 128, 128)

[CatalinVulcan]

I'm trying to use the pre-trained model for prediction using the following command:

python dsen2cr_main.py --predict model_SARcarl.hdf5

But I'm getting the following error. Do you have any idea why this is?

Traceback (most recent call last):
File "dsen2cr_main.py", line 198, in
run_dsen2cr(args.predict_file, args.resume_file)
File "dsen2cr_main.py", line 145, in run_dsen2cr
include_sar_input=include_sar_input)
File "C:\Users\Catalin\Desktop\cloud removal\dsen2cr_network.py", line 52, in DSen2CR_model
x = Add()([x, input_opt])
File "C:\Users\Catalin\Soft\anaconda3\dsen2cr_env\lib\site-packages\tensorflow_core\python\keras\engine\base_layer.py", line 824, in call
self._maybe_build(inputs)
File "C:\Users\Catalin\Soft\anaconda3\dsen2cr_env\lib\site-packages\tensorflow_core\python\keras\engine\base_layer.py", line 2146, in _maybe_build
self.build(input_shapes)
File "C:\Users\Catalin\Soft\anaconda3\dsen2cr_env\lib\site-packages\tensorflow_core\python\keras\utils\tf_utils.py", line 306, in wrapper
output_shape = fn(instance, input_shape)
File "C:\Users\Catalin\Soft\anaconda3\dsen2cr_env\lib\site-packages\tensorflow_core\python\keras\layers\merge.py", line 111, in build
output_shape = self._compute_elemwise_op_output_shape(output_shape, shape)
File "C:\Users\Catalin\Soft\anaconda3\dsen2cr_env\lib\site-packages\tensorflow_core\python\keras\layers\merge.py", line 82, in _compute_elemwise_op_output_shape
'together with shapes ' + str(shape1) + ' ' + str(shape2))
ValueError: Operands could not be broadcast together with shapes (13, 128, 13) (13, 128, 128)

[ameraner]

Hi @CatalinVulcan . Thanks for reporting the issue. I would suspect an issue where the dimensions are in the wrong order (the bands dimension being the third dimension instead of the first). Could you check the input shapes that you pass to the model

dsen2-cr/Code/dsen2cr_main.py

Line 121 in 59e81ae

model, shape_n = DSen2CR_model(input_shape,

?
Is this line also still active in your code:

dsen2-cr/Code/dsen2cr_main.py

Line 16 in 59e81ae

K.set_image_data_format('channels_first')

?

[CatalinVulcan]

Thank you for your reply. It had to do with the version of Numpy. But in the end, I couldn't make it work due to library versioning and compatibility.

Using

conda install -c conda-forge python=3.7 tensorflow-gpu=1.15.0 keras=2.2.4 numpy scipy rasterio pydot graphviz h5py,

resulted in installing some package versions that gave errors when running the code. I tried changing the versions, but every time I changed one, other errors and incompatibilities appeared until I reached a point I simply couldn't solve.

For input, I used images from the dataset mentioned in the README. It might help to include 3 images in the repository as well, along with the appropriate datasetfilelist.csv. This might make things simpler.

[ameraner]

Ah, yeah. Unfortunately I have no means atm to try to update the code for newer package versions. If you ever get it running, feel free to open a PR or propose changes! Also, good point about the images..

[ameraner]

@animecomico has reported in #6 (comment) that this works:

conda create -y -n dsen2_cr_v1 python=3.7.0
conda activate dsen2_cr_v1
conda install numpy=1.17.5
pip install tensorflow==1.13.2
pip install keras==2.2.4
pip install matplotlib==2.1.1
pip install rasterio
pip install pydot
pip install h5py
pip install graphviz
pip install scipy
pip install protobuf==3.6.1