ValueError: Only sparse matrices with 32-bit integer indices are accepted. Got int64 indices

torchx
1

when training svm by sklearn with fashion_mnist dataset .
ValueError: Only sparse matrices with 32-bit integer indices are accepted. Got int64 indices.
Can you help me? Thanks!

2

It seems you are running into a scikit-learn error, which doesn’t seem to be PyTorch-related, so you would get a better and faster answer in a scikit-specific board.
In any case, check the dtype of the input values to this classifier, compare it to the docs, and make sure they are valid.

1 Like
3

import hub

ds = hub.load(‘hub://activeloop/caltech101’)

dataloader = ds.pytorch(num_workers=0, batch_size=4, shuffle=False)

hi, i want to train model with this dataset.but how to i split this data set into training set and test set?