Puzzles

Split the iris dataset into training and test sets with 80% for training and 20% for testing. Use random_state=42 for reproducibility.

Train a K-Nearest Neighbors classifier on the iris dataset and make predictions on the test set.

Create and display a confusion matrix for a classifier's predictions on the iris dataset.

Use StandardScaler to standardize features so they have zero mean and unit variance.

Use 5-fold cross-validation to evaluate a Random Forest classifier on the digits dataset.

Create a pipeline that first scales the data and then applies logistic regression.

Use GridSearchCV to find the best hyperparameters for an SVM classifier.

Train a Random Forest and extract the feature importances for the wine dataset.

Use PCA to reduce the digits dataset from 64 dimensions to 2 and visualize explained variance.

Apply K-Means clustering to the iris dataset and compare cluster labels with true labels.

Train a classifier and compute the ROC curve and AUC score for binary classification.

Create a voting classifier that combines multiple different classifiers.

Calculate precision, recall, and F1-score for different classification thresholds.

Use ColumnTransformer to handle numerical and categorical features differently in a pipeline.