0
votes

I have the following code, I am trying to receive predictions through the TCP socket. While doing so I am able to receive the data but I retrieve error

while True:
    frame = footage_socket.recv_pyobj()
    print(type(frame)) # <class 'numpy.ndarray'>
    predictions = img_to_array(frame)
    tmp = np.zeros( predictions.shape )
    for i in range( 0, 1 ):
        tmp[i,:] = predictions[i, :]
        predictions_result = m2.predict( tmp )
        label_vgg16 = decode_predictions( predictions_result )
 footage_socket.close()

Error

predictions = img_to_array(frame) ValueError: Unsupported image shape: (1, 224, 224, 64)

Thanks, help is highly appreciated.

If you already have a numpy array, there is no need to use img_to_array - Dr. Snoopy
Thanks alot for your wonderful help you saved my day @Dr.Snoopy - Mazia