Join Our 5-Week ML/AI Engineer Interview Bootcamp 🚀 led by ML Tech Leads at FAANGs
Back to Questions
77. Precision And Recall
easy
Applesenior
Using the appstore_transaction_risk DataFrame, what are the precision and recall when classifying transactions as positive if risk_score > 0.80, with true_label = 1 as the positive class? Return precision_score and recall_score rounded to 2 decimal places.
appstore_transaction_risk
| Column Name | Type |
|---|---|
| transaction_id | int64 |
| risk_score | float64 |
| true_label | int64 |
Expected Output Schema
| Column Name | Type |
|---|---|
| precision_score | float64 |
| recall_score | float64 |
Constraints
Use DataFrame columns: risk_score, true_label.
Threshold strictly greater than 0.80.
Round precision/recall to 2 decimals.
Hint 1
Create a binary prediction column: pred = (risk_score > 0.80).astype(int).
Hint 2
Compute confusion counts with boolean masks: TP, FP, FN using pred vs true_label.
Hint 3
Use precision = TP/(TP+FP) and recall = TP/(TP+FN); handle zero denominators; round to 2 decimals.
Roles
Data Scientist
Data Analyst
Data Engineer
Companies
AppleLevels
senior
entry
Tags
pandas
classification-metrics
precision-recall
thresholding
confusion-matrix
43 people are solving this problem
Python EditorLn 1, Col 1