kkgadiraju/multi-modal-crop-classification

Code for the paper Multi Modal Deep Learning Based Crop Classification Using Multispectral and Multitemporal Satellite Imagery published in KDD Applied Data Science Track 2020

multi-modal-crop-classification's Issues

data points

Thank you for your remarkable work. I would like to inquire about the principle of data point selection in your paper. I noticed that you mentioned collecting over 60,000 data points, and I'm curious about how you determined these points. Was it simply based on intuition, or are the coordinates of these points the center of counties? This point has puzzled me, although you wrote in your paper about selecting based on the major crops of the states.

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

Hello, your paper on the multi-modal method on crop classification is really great.

But when I followed your readme, renamed data and result folders' names, then tried to replicate your result using "train-gridsearch.py" in folders lstm or bilstm, I encountered the error message of "ValueError: Tried to convert 'y' to a tensor and failed. Error: None values not supported."

When I dug into the codes, I found it happened when it called line 92 in "optimziers.py": "None in grads". I am wondering whether you have any encounter this issue or have you got any suggestion for me?

Thank you very much for your reading and help.

Error While Running the concatenation model

I was running the concatenation module and i did not alter any model layer values but still facing the error:
File "train-gridsearch.py", line 203, in
history = model.fit(train_generator,
File "/usr/local/lib/python3.8/dist-packages/tensorflow/python/keras/engine/training.py", line 1187, in fit
tmp_logs = self.train_function(iterator)
File "/usr/local/lib/python3.8/dist-packages/tensorflow/python/util/traceback_utils.py", line 153, in error_handler
raise e.with_traceback(filtered_tb) from None
File "/tmp/autograph_generated_file52u8rjgw.py", line 15, in tf__train_function
retval = ag_.converted_call(ag__.ld(step_function), (ag__.ld(self), ag__.ld(iterator)), None, fscope)
File "/usr/local/lib/python3.8/dist-packages/keras/utils/traceback_utils.py", line 70, in error_handler
raise e.with_traceback(filtered_tb) from None
File "/usr/local/lib/python3.8/dist-packages/keras/engine/input_spec.py", line 232, in assert_input_compatibility
raise ValueError(
ValueError: in user code:

File "/usr/local/lib/python3.8/dist-packages/keras/utils/traceback_utils.py", line 70, in error_handler
    raise e.with_traceback(filtered_tb) from None
File "/usr/local/lib/python3.8/dist-packages/keras/engine/input_spec.py", line 232, in assert_input_compatibility
    raise ValueError(

ValueError: Input 0 of layer "zero_padding2d" is incompatible with the layer: expected ndim=4, found ndim=2. Full shape received: (None, 1)

Can u please help why it is showing this behavior as model layer values are alright