After fitting the model the R2 for test data is negative

Can we use sklearn R2 to compare the performance on out of sample dataset? Or is MAPE more reliable here to use?