One hot encoding in pytorch

Question

I am really new to coding, right now I am trying to turn my label to one hot encoding. I have already done transferring the np.array to tensor as shown below

tensor([4., 4., 4., 4., 4., 4., 4., 4., 4., 4., 4., 4., 4., 4., 4., 1., 1., 1.,
        1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 3., 3., 3., 3., 3., 3.,
        3., 3., 3., 3., 3., 3., 3., 3., 3., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
        0., 0., 0., 0., 0., 0., 2., 2., 2., 2., 2., 2., 2., 2., 2., 2., 2., 2.,
        2., 2., 2.], dtype=torch.float64)

and I am using code to do one hot encoding 

aaa = F.one_hot(torch_qyh, num_classes=5)

However, there is an error showing "RuntimeError: one_hot is only applicable to index tensor." Any help would appreciated.

Answer 1

You will have to convert it in long type. Can't do it with float. F.one_hot only takes LongTensor.

F.one_hot(t.long())