Python/ Expected axis -1 of input shape to have value 3 but received input with shape (None, 32, 32, 1)

[Zuccers]

 

Hello, I made an image recognition model in python using keras, this is the input shape: 

input_shape=(3, 32, 32)

The model itself works fine, but when I try to pass any other image that is not from the dataset I get this error:

ValueError: Input 0 of layer sequential is incompatible with the layer: expected axis -1 of input shape to have value 3 but received input with shape (None, 32, 32, 1)

I do reshape the image, but can not figure out why I am getting that error and failing every pass

image = load_img("image.jpeg", grayscale="false", color_mode="rgb", target_size=(32, 32))
image = img_to_array(image)
image = image.reshape(3, 32, 32, 1)
image = image.astype('float32')
image = image / 255.0

How do I go about solving this issue and reshaping the image correctly?

Thanks in advance!

[Sauron]

please post the image and the whole code as well as the entire output of the program so we can see the context of the error.

[igormp]

On 1/7/2021 at 7:36 AM, Zuccers said:

 

Hello, I made an image recognition model in python using keras, this is the input shape: 

input_shape=(3, 32, 32)

The model itself works fine, but when I try to pass any other image that is not from the dataset I get this error:

ValueError: Input 0 of layer sequential is incompatible with the layer: expected axis -1 of input shape to have value 3 but received input with shape (None, 32, 32, 1)

I do reshape the image, but can not figure out why I am getting that error and failing every pass

image = load_img("image.jpeg", grayscale="false", color_mode="rgb", target_size=(32, 32))
image = img_to_array(image)
image = image.reshape(3, 32, 32, 1)
image = image.astype('float32')
image = image / 255.0

How do I go about solving this issue and reshaping the image correctly?

Thanks in advance!

image = image.reshape(3, 32, 32, 1)

Should be:

image = image.reshape(3, 32, 32)