Logistic Regression

$*$in python

Objective: Implement Logistic Regression from scratch to classify the Iris dataset and compute the True Positive, False Positive, True Negative, False Negative rates.

$1.$ Load and Preprocess the Iris Dataset

$-$ Load the Iris dataset from sklearn.datasets.

$-$Extract only the first two classes to make it a binary classification problem.

$-$ Split the dataset into features $($X$)$ and target labels $($y$).$

$-$Split the dataset into training and testing sets using train test split with test$\_$size$=0\mathrm{.}2$ and random state$=42.$

$-$Use the following code for feature scaling.

$-$Standardize the features using:

StandardScaler from sklearn.preprocessing.scaler$=$StandardScaler$()$

Xtrain $=$ scaler.fit transform$($X train$)$ X$\_$test $=$scaler.transform$($X$\_$test$)$

$2.$ Model Training

$-$Train a Logistic Regression model using LogisticRegression from sklearn.linear$\_$model on the training data.

$-$Keep the default parameters.

$3.$ Make Predictions

$-$Predict the classes and print them.

$-$Manually calculate and print the True Positives$($TP$),$ True Negatives $($TN$),$ False Positives$($FP$),$ and False Negatives$($FN$)$ using the actual and predicted classes.

$-$Manually calculate and print and Root Mean Squared Error$($RMSE$).$