TypeError: can't convert CUDA tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first. about ganomaly

[samet-akcay]

@yoheikikuta, Thanks for your feedback, the repo has now been updated.

[yhxuyou]

hello,I have done this,but i got an another problem like this,and how can i to deal with it?

Traceback (most recent call last):
  File "train.py", line 43, in <module>
    model.train()
  File "/data4/xyz/genarmal/lib/model.py", line 280, in train
    res = self.test()
  File "/data4/xyz/genarmal/lib/model.py", line 354, in test
    auc = evaluate(self.gt_labels, self.an_scores, metric=self.opt.metric)
  File "/data4/xyz/genarmal/lib/evaluate.py", line 30, in evaluate
    return roc(labels, scores)
  File "/data4/xyz/genarmal/lib/evaluate.py", line 55, in roc
    fpr, tpr, _ = roc_curve(labels, scores)
  File "/usr/local/lib/python3.6/site-packages/sklearn/metrics/ranking.py", line 534, in roc_curve
    y_true, y_score, pos_label=pos_label, sample_weight=sample_weight)
  File "/usr/local/lib/python3.6/site-packages/sklearn/metrics/ranking.py", line 318, in _binary_clf_curve
    raise ValueError("{0} format is not supported".format(y_type))
ValueError: multiclass format is not supported

[samet-akcay]

@yhxuyou, I'm not sure, but your problem sounds more like that your data. It sounds like your data is a multi-class dataset, which would not work in this case since this repo only accepts binary class at the moment.

[yoheikikuta]

Since the original issue was solved, I'll close the issue.
Thanks.