In binary and multilabel cases, the elements of y and y_pred should have 0 or 1 values. Thresholding of predictions can be done as below: def thresholded_output_transform(output): y_pred, y = output y_pred = torch.round(y_pred) return y_pred, y metric = Accuracy(output_transform=thresholded_output_transform) metric.attach(default_evaluator .... "/> knitting patterns for cotton aran inexpensive birthday party ideas