Chuanlei Ni
internal_df = self._df.pivot(group_by=col1, pivot_col=col2, value_col=daft.lit(0), agg_fn="count")
jay