T >= 0 && t < n_classes error

vision
1

I’m trying to load EMNIST from pytorch datasets which have 26 classes.
So I modify vgg16() model to have 26 num_classes (1 to 26). but this error occurs.
What`s wrong?

/pytorch/aten/src/THCUNN/ClassNLLCriterion.cu:105: void cunn_ClassNLLCriterion_updateOutput_kernel(Dtype *, Dtype *, Dtype *, long *, Dtype *, int, int, int, int, long) [with Dtype = float, Acctype = float]: block: [0,0,0], thread: [3,0,0] Assertion `t >= 0 && t < n_classes` failed.
2

Print the targets coming out of your dataset and make sure they are within [0,25]. Are they?