Running text_classification_introspective_rationale_explainer.ipynb with multi-class dataset

[HenryLiuPI]

I'm getting RuntimeError: multi-target not supported at ..\aten\src\THNN/generic/ClassNLLCriterion.c:20 while trying to feed multi-class dataset.

My training data looks like this. I'm using one-hot format as label.

I'm getting the error on explainer.fit(df_train, df_test)

The dataset is a multi-class dataset and has 8 labels in total. Each sample may contain multi labels. What format should I use to represent the label?