Could you check the weight
and bias
in both layers?
Sometimes, e.g. when the learning rate is too high, the model just learns the “mean prediction”, i.e. the bias
is responsible for most of the prediction, while the weight
s and input
became more or less useless.
For example when I was playing with a facial keypoint dataset, some models just predicted the “mean position” of the keypoints, regardless of the input image.
6 Likes