• The code for each Part of this should be self-contained, that is, each of Part 1, 2, and 3 should contain all the necessary code and not rely on code from another Part of the lab in order to run.
• all parts of the lab should be done using python, sklearn, pandas, numpy, and matplotlib.

Part 1 - Creating and evaluating a random forest model

In this part of the lab, you should:

• read in the data;
• verify that all the data is numeric and that there are no missing values;
• split the data into training and validation sets (don't worry about creating a final test set);
• create a random forest model using the data;
• evaluate the model on both the training and validation sets using MAE and % error.

Part 2 - Exploring the n_estimators hyper-parameter

In this part of the lab you should:

• use a for loop to create a random forest model for each value of n_estimators from 1 to 30;
• evaluate each model on both the training and validation sets using MAE;
• visualize the results by creating a plot of n_estimators vs MAE for both the training and validation sets.

After that you should answer the following questions:

• Which value of n_estimators gives the best results?
• Explain how you decided that this value for n_estimators gave the best results;
• Why is the plot you created above not smooth?
• Was the result here better than the result of Part 1? What % better or worse was it?

Part 3 - Exploring the max_features hyper-parameter

In this part of the lab you should:

• use a for loop to create a random forest model for each value of max_features from 1 to the total number of features in the data;
• for each model, use the value for n_estimators as determined in Part 2;
• evaluate each model on both the training and validation sets using MAE;
• visualize the results by creating a plot of max_features vs MAE for both the training and validation sets.

After that you should answer the following questions:

• Which value of max_features gives the best results?
• Explain how you decided that this value for max_features gave the best results;
• Was the result here better than the result of Part 2? What % better or worse was it?