Hyper-parameter tuning and model persistence done

[khalidcawl]

• hyper-parameter tuning done
• saving cross validation plot done
• fixed out_dir argument in script
• to access the model object, do this:

  from joblib import dump, load
  pipe_loaded = load('best_model_pipe.joblib')
  pipe_loaded.named_steps['ridge'].coef_