Logistic Regression Implementation | Dataford Interview Questions

Tell me about how you would implement a logistic regression model from data, and then iterate on it to improve performance.

Implement your answer as a pure function that trains binary logistic regression using batch gradient descent and returns predicted labels for the training rows. Include a learning rate, epoch count, and L2 regularization parameter.

Input X is a non-empty list of numeric feature rows and y is a binary-label list with matching length. Return a list of integer predictions in the same order as X.

Constraints

1 <= len(X) <= 500
1 <= len(X[i]) <= 50
len(X) == len(y)
Each feature value is a finite number
Each label in y is either 0 or 1
learning_rate > 0
epochs >= 1
l2 >= 0

Function Signature

def logistic_regression(X, y, learning_rate, epochs, l2):

Your question is Logistic Regression Implementation. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.

Problem

Tell me about how you would implement a logistic regression model from data, and then iterate on it to improve performance.

Input X is a non-empty list of numeric feature rows and y is a binary-label list with matching length. Return a list of integer predictions in the same order as X.

Constraints

1 <= len(X) <= 500
1 <= len(X[i]) <= 50
len(X) == len(y)
Each feature value is a finite number
Each label in y is either 0 or 1
learning_rate > 0
epochs >= 1
l2 >= 0

Function Signature

def logistic_regression(X, y, learning_rate, epochs, l2):

Your question is Logistic Regression Implementation. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.