Part $1$: Data Preparation and Feature Engineering

Load the California Housing dataset.

Handle Missing Data using an appropriate imputation technique to handle any missing values.

Feature Encoding: for any ordinal categorical variables, apply Label Encoding; for nominal categorical variables, use One$-$Hot Encoding.

Normalize or standardize the numerical features.

Part $2$: Dimensionality Reduction with PCA

Calculate the covariance matrix of the scaled data and find its eigenvalues and eigenvectors.

Use the eigenvalues to compute the principal components.

Select the number of principal components that preserve at least $90\%$ of the variance.

Part $3$: Model Training and Evaluation

Train a simple classification model $($such as logistic regression or decision tree$)$ on the original dataset $($before applying PCA$).$

Train the same classification model on the dataset after applying PCA.

Compare the performance of the baseline model and the PCA$-$transformed model using accuracy, precision, recall, and F$1-$score.

THERE IS california$\_$housing$\_$test.csv and california$\_$housing$\_$train.csv