How to work with a simple dataset that doesn't involve a .mat file?

[Harsh9524]

I have a medical dataset of Lung X-Rays of Pneumonia patients and I want to apply this code on that dataset.
The dataset is in the following format.
Pneumonia/
----img1.jpg
----img2.jpg
Normal/
----img1.jpg
----img2.jpg

[changspencer]

You already have them in a picture form, so I would start with understanding the prepare_data method in kfold.py. It probably just needs some quick edits on being used for JPG images and depends on if you have a mask for it or not.

My memory is a little fuzzy, so if you still have questions about the code after looking for a while, let me know.

[Harsh9524]

Hi @changspencer ,

Thank you for your response.
As per your suggestion I tried using my dataset directly. and since I have a preprocess.py file which outputs x_train.npy, y_train.npy, x_test.npy and y_test.npy I removed the prepare_data function from the kfold.py file and directly inputted my data in the form of numpy arrays. Since the prepare_data function was also returning numpy arrays.

After doing this and running the kfold.py file I encounter the following error. Can you please have a look at it?

Traceback (most recent call last):
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/ops.py", line 1864, in _create_c_op
    c_op = c_api.TF_FinishOperation(op_desc)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Dimensions must be equal, but are 3 and 2 for 'mask_1/mul' (op: 'Mul') with input shapes: [?,3,16], [?,2,1].

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "kfold.py", line 278, in <module>
    main()
  File "kfold.py", line 273, in main
    num_folds, sep_conv, use_margin_loss)
  File "kfold.py", line 196, in k_fold_validation
    model = build_model(use_margin)
  File "kfold.py", line 91, in build_model
    masked_by_y = Mask()([capLayer2, y])
  File "/usr/local/lib/python3.6/dist-packages/keras/engine/base_layer.py", line 457, in __call__
    output = self.call(inputs, **kwargs)
  File "/content/Tumor-CapsNet/capsulelayers.py", line 61, in call
    masked = K.batch_flatten(inputs * K.expand_dims(mask, -1))
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/math_ops.py", line 884, in binary_op_wrapper
    return func(x, y, name=name)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/math_ops.py", line 1180, in _mul_dispatch
    return gen_math_ops.mul(x, y, name=name)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/gen_math_ops.py", line 6490, in mul
    "Mul", x=x, y=y, name=name)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/op_def_library.py", line 788, in _apply_op_helper
    op_def=op_def)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/util/deprecation.py", line 507, in new_func
    return func(*args, **kwargs)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/ops.py", line 3616, in create_op
    op_def=op_def)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/ops.py", line 2027, in __init__
    control_input_ops)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/ops.py", line 1867, in _create_c_op
    raise ValueError(str(e))
ValueError: Dimensions must be equal, but are 3 and 2 for 'mask_1/mul' (op: 'Mul') with input shapes: [?,3,16], [?,2,1].

[Harsh9524]

@changspencer
UPDATE: I fixed the error by adjusting the layer 3. and changing the num_capsule from 3 to 2 since I have only two classes in my dataset.

CapsuleLayer(num_capsule=2, dim_capsule=16, routings=2,
                             name="ThirdLayer")(primaryCaps)