shahinshh/covid-fact

A Fully-Automated Capsule Network-based Framework for Identification of COVID-19 Cases from Chest CT scans

covid-fact's Issues

Reshape error

Hi and thank you for your code!
I am trying to run the indentification script, but running into a reshape error. My test input is two DICOM file of a lung scan with the shape of 512x512. The error code I am running into is

tensorflow.python.framework.errors_impl.InvalidArgumentError:  Input to reshape is a tensor with 1048576 values, but the requested shape has 32768
	 [[node functional_1/capsule_1/Reshape (defined at COVID-FACT_binary_test.py:123) ]] [Op:__inference_predict_function_1338]

Errors may have originated from an input operation.
Input Source operations connected to node functional_1/capsule_1/Reshape:
 functional_1/capsule_1/convolution/Squeeze (defined at COVID-FACT_binary_test.py:114)

Function call stack:
predict_function

About training code

Thanks a lot for sharing code! I've read the papers about the dataset and model. But I'm still confused about how to use the unlabeled subset to develop the second stage of the model. May I get the training code? Thanks again!
[email protected]

Error in running COVID-FACT_binary_test.py

Hi everyone,
I have installed COVID-FACT on Ubuntu 18
when I was running the python script I got the following message:

Using TensorFlow backend.
Illegal instruction (core dumped)

Any advice ?
Thanks
Jaafar

<class 'tuple'> to Tensor Error

First of all, I want to thank you for the code provided and the effort put into it.
I have encountered the reshape problem that was already raised as an issue, so I followed your proposed fix and downgraded the version of keras and tensorflow to those mentioned in the README (tensorflow 1.14.0 and keras 2.2.4), however another problem was raised when instantiating the 'Capsule' class, the problem is as follows:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/tensor_util.py in make_tensor_proto(values, dtype, shape, verify_shape, allow_broadcast)
    557     try:
--> 558       str_values = [compat.as_bytes(x) for x in proto_values]
    559     except TypeError:

26 frames
TypeError: Expected binary or unicode string, got 1

During handling of the above exception, another exception occurred:

TypeError                                 Traceback (most recent call last)
/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/tensor_util.py in make_tensor_proto(values, dtype, shape, verify_shape, allow_broadcast)
    560       raise TypeError("Failed to convert object of type %s to Tensor. "
    561                       "Contents: %s. Consider casting elements to a "
--> 562                       "supported type." % (type(values), values))
    563     tensor_proto.string_val.extend(str_values)
    564     return tensor_proto

TypeError: Failed to convert object of type <class 'tuple'> to Tensor. Contents: (1, Dimension(128), 512). Consider casting elements to a supported type.