Reference no: EM132173839

Using R Programming, Implement Logistic Regression using Gradient Descent.

You can use any gradient descent technique (batch or stochastic).

Choose any set of initial parameters for gradient descent.

Choose any convergence criteria.

Try to vectorize code by avoiding for loops.

Then, using the code created above,

a) Train 2 models, one on the train_0_1 set and another on train_3_5, and report the training and test accuracies.

b. Repeat 3a 10 times, i.e. you should obtain 10 train and test accuracies for each set. Calculate the average train and test accuracies over the 10 runs, and report them.

c. For 0,1 and 3,5 cases, explain if you observe any difference you in accuracy. Also, explain why do you think this difference might be.

d. This assignment deals with binary classification. Explain what you would do if you had more than two classes to classify, using logistic regression.

Provide:

R code of a predict function to calculate the predictions, accuracy function to calculate accuracy, calls to these function for a) & b) above

For a) , train and test accuracies for 0/1 and 3/ 5 classification.

For b), average test and train accuracies for 0/1 and 3/5 classification.