What does it mean to underfit your data model?
Your model has low bias but high variance.
There is not a lot of variance but there is a high bias.
There is too much data in your training set.
There is too little data in your training set.