Next steps for Random Forest (input from Professor Santerre)

[jtfields]

1. Covariance analysis
2. Breakpoint of 15% for cold
3. Plot hist with 20 columns
4. Go to 4-5 categories
5. Regression on meaningful values (continuous)
6. Look at just familarity and hotttnesss
7. XG boost will speed up RF
8. Review lines 49-58 in ml approaches and try other techniques