Am I using onehot-encoded features correctly in the Machine-Learning model?

Question

I have 14 features and 1 label for my RenadomForestRegressor model by using Sklearn-packages in Python. I have 10000 data under each column, so the si