Thank you for writing and maintaining this repo, it's a very easy to understand implementation of Score-CAM!
I have the VGG16 example working fine, and am now trying to adapt this to use my own Mask-RCNN model I have developed using this repo as a basis. I am running into an error however that I am unsure of how to fix, and was wondering if any advice could be provided?
explainer = ScoreCAM()
target_size = (3456, 5184)
data = (img, None)
# Compute ScoreCAM on VGG16
image = explainer.explain(
data, model.keras_model, max_prob, target_size, _grid=False
)[0]
explainer.save(image, ".", "score_cam.png")
ValueError: Error when checking model input: the list of Numpy arrays that you are passing to your model is not the size the model expected. Expected to see 3 array(s), but instead got the following list of 1 arrays: [<tf.Tensor: id=10530, shape=(64, 3456, 5184, 3), dtype=float32, numpy=
array([[[[ 3.64952057e+02, 3.75305298e+02, 3.77893616e+02],
[ 1.37095261e+02, 1.40957092e+02, 1.41922562e+02],