Medium

Logistic Classifier Predictions

Medium

~15 min

code completion

Logistic Regression Prediction

Logistic regression converts a linear score into a binary prediction:

1. Compute the linear combination: $z = X w$

2. Apply sigmoid: $\overset{p}{^} = σ (z)$

3. Threshold: $\overset{y}{^}_{i} = 1 [\overset{p}{^}_{i} \geq threshold]$

The default threshold is 0.5, but it can be adjusted to trade precision vs. recall.

Your task:

Implement logistic_predict(X, weights, threshold) that returns an integer array of 0s and 1s.

Example Tests

Zero input: sigmoid=0.5, above default threshold

Input: {"X":[[0],[10],[-10]],"weights":[1],"threshold":0.5}

Expected: [1,1,0]

Negative scores: all predict 0

Input: {"X":[[1,2],[3,4]],"weights":[0.5,-1],"threshold":0.5}

Expected: [0,0]

Raised threshold makes prediction stricter

Input: {"X":[[0],[0],[10]],"weights":[1],"threshold":0.8}

Expected: [0,0,1]

You can read the full problem statement above. Create a free account to run code in the browser, submit solutions, and track your progress.