Scribd
Upload
12 views3 pages

Peer Review Guide for Explainable AI

The document outlines the Peer Review Assignment process for a DPhi Mini Bootcamp on Explainable AI, emphasizing the importance of respectful and constructive feedback among learners. It details the scoring rubric across multiple tasks, including data analysis, model prediction, and SHAP value interpretation, with a total of 70 marks available. Additionally, it encourages learners to explore SHAP further for potential bonus points.

Uploaded by

John Dale Vacaro
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
12 views3 pages

Peer Review Guide for Explainable AI

The document outlines the Peer Review Assignment process for a DPhi Mini Bootcamp on Explainable AI, emphasizing the importance of respectful and constructive feedback among learners. It details the scoring rubric across multiple tasks, including data analysis, model prediction, and SHAP value interpretation, with a total of 70 marks available. Additionally, it encourages learners to explore SHAP further for potential bonus points.

Uploaded by

John Dale Vacaro
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Welcome to Peer Review Assignments

DPhi Mini Bootcamp: Explainable AI

A Peer Review Assignment is where you and your fellow learners have an opportunity to
review and grade each other’s work. You can provide both qualitative and quantitative feedback
based on a rubric that we have created (available at the end of this document). Peer reviews
also provide pedagogical benefits and can help learners develop evaluative judgement and self-
regulation.

Expectations: Reviewers

While grading, be respectful, encouraging, and honest. Acknowledge what your fellow learner
did well and offer specific suggestions on how they can improve. Scores should reflect the
learner’s grasp of the concepts and should not be deducted points for difficulties with language
or differences of opinion.

Note: Reviewing peers assignment carries 20 marks. Your reviewed assignments’


feedback and scores will be reviewed by the DPhi team. Any anomalies in the scores will
carry a penalty. We request all the learners to review notebooks responsibly.
Scoring Rubric
You can look at the solution notebook here:
[Link]
usp=sharing

Task 1:
1. Import Necessary Libraries
2. Load the Dataset - Display the first 5 rows of the data
3. Perform Basic Exploratory Data Analysis (EDA)
4. Split the data into Train and Test Sets
○ The learner has correctly done all four - 10 marks

Task 2:
1. Use Random Forest Machine Learning Model for prediction
○ The learner has correctly used random forest model for prediction - 5 mark
2. Evaluate the model using Accuracy Score
○ The learner has evaluated and printed the accuracy score - 5 mark

Task 3: Use a SHAP Explainer to derive SHAP Values for the random forest ml model.
○ The learner has used shap explainer to derive and print the values - 5 mark
○ The learner has written the inferences - 5 mark

Task 4: Plot a SHAP force plot for the first row of test data.
○ The learner has plotted the shap force plot on the first row of data - 5 mark
○ The learner has written the inferences from the plot - 5 mark

Task 5: Plot a SHAP force plot for all the rows of the data
○ The learner has plotted the shap force plot on all rows of data - 5 mark
○ The learner has written the inferences from the plot - 5 mark

Task 6: Plot a SHAP summary plot using all the features in the data
○ The learner has plotted the shap summary plot for all features of data - 5 mark
○ The learner has written the inferences from the plot - 5 mark

Task 7: Plot a SHAP dependence plot using all features in the data
○ The learner has plotted the shap dependence plot for all or any features of data -
5 mark
○ The learner has written the inferences from the plot - 5 mark
Task 8 (Optional): Do anything and everything with SHAP
○ If the learner has done anything else to explore shap, please feel free to give
bonus points (do not exceed the maximum marks while doing this)

Final marks to be submitted (out of 70): Sum of all obtained marks + bonus
marks for Task 8(if any).

You might also like