assess your course 6 end-of-course project coursera answers

This is the rubric for the Course 6 end-of-course project. You will use this rubric to review and grade your own work. The rubric grading process is an important part of the learning experience because it allows you to objectively assess your end-of-course project against a set of criteria.

There are a total of 24 points for the end-of-course project and 24 items in this rubric. Each rubric item is worth 1 point. The items are grouped by topic and correspond to each step you completed for the Course 6 end-of-course project. 

To use the rubric, first open your end-of-course project notebook, executive summary, and PACE strategy document. Next, review each rubric item’s grading criteria. Then respond to each statement by marking “yes” or “no.” 

When you complete and submit the rubric, you will receive a percentage score. This score will help you confirm whether you completed the required steps of the end-of-course project; the recommended passing grade for this project is 80% (or 19/24 points). If you want to increase your score, you can revise your project and then resubmit this rubric to reflect any changes you make. Try to achieve at least 19 points on this rubric before continuing on to the next course.

Imports

The following rubric items assess the imports for your end-of-course project.

1. Applicable packages and libraries were imported to the code notebook.

  • Yes
  • No

2. The dataset was imported and read into the notebook using the pd.read_csv() function.

  • Yes
  • No

3. All introductory questions in the notebook were answered.

  • Yes
  • No

Feature Engineering

The following rubric items assess the feature engineering work you completed for your end-of-course project.

4. A dataframe was created that contains only customers who paid with credit cards.

  • Yes
  • No

5. A new column that calculates tip percentage based on ‘tip_amount’ and ‘total_amount’ columns was added.

  • Yes
  • No

6. A binary column was created that indicates whether a customer tipped ≥ 20%.

  • Yes
  • No

7. A new column for trip duration was created by subtracting dropoff time from pick up time.

  • Yes
  • No

8. A column that names the day of the week the ride occurred was created.

  • Yes
  • No

9. Four more columns were created categorizing pick up times into four categories—AM rush, daytime, PM rush, and nighttime—and each value in all four new columns was converted to binary.

  • Yes
  • No

10. Categorical variables were encoded as binary dummy variables.

  • Yes
  • No

11. After all columns were created and converted, the following columns were dropped: payment_type, mta_tax, tip_amount, total_amount, tip_percent.

  • Yes
  • No

12. “Precision” was the evaluation metric identified that assesses true/false positives; in this case, best for taxi drivers.

  • Yes
  • No

Machine Learning Modeling

The following rubric items assess the machine learning modeling you completed for your end-of-course project.

13. The data was split into predictor variables and target variables, with target being the binary indicator of whether customers tipped ≥ 20%.

  • Yes
  • No

14. The predictor variables and target variables were split into train and test sets.

  • Yes
  • No

15. The following steps were performed for the random forest model:

  • Performed a GridSearch to tune hyperparameters
  • Captured precision, recall, F1 score, and accuracy metrics
  • Refit the model with best average precision score across all validation folds
  • Obtained validation scores of best model
  • Scored the model on the test data

16. The following steps were performed for the XGBoost model:

  • Performed a GridSearch to tune hyperparameters
  • Captured precision, recall, F1 score, and accuracy metrics
  • Refit the model with best average precision score across all validation folds
  • Obtained validation scores of best model
  • Scored the model on the test data

17. The best XGBoost model was compared to the best random forest model.

  • Yes
  • No

18. A confusion matrix of the XGBoost model’s prediction results was plotted.

  • Yes
  • No

19. The plot_importance() function was used to identify the top 10 most important predicting features of the XGBoost model.

  • Yes
  • No

Results and/or Evaluation 

The following rubric items assess the concluding steps of your end-of-course project, including evaluation and summary of findings.

20. All questions in the code notebook were answered.

  • Yes
  • No

21. All questions in the PACE strategy document were answered.

  • Yes
  • No

22. The executive summary clearly articulated the challenges presented in this data project.

  • Yes
  • No

23. The executive summary identified the outcome of your work.

  • Yes
  • No

24. The executive summary included recommendations for future work/next steps.

  • Yes
  • No

Leave a Reply