Pipeline with GridSearchCV in ML Python - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a pipeline with a scaler and a logistic regression model.

ML Python

from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression

pipeline = Pipeline([('scaler', StandardScaler()), ('clf', [1])])

Drag options to blanks, or click blank then click option'

AKNeighborsClassifier()

BRandomForestClassifier()

CLogisticRegression()

DSVC()

Attempts:

3 left

💡 Hint

Common Mistakes

Using a classifier other than LogisticRegression when the task specifies logistic regression.

Forgetting to instantiate the model with parentheses.

The pipeline requires a logistic regression model instance as the classifier step.

2fill in blank

medium

Complete the code to define the parameter grid for GridSearchCV to tune logistic regression's C parameter.

ML Python

param_grid = {'clf__[1]': [0.1, 1, 10]}

Drag options to blanks, or click blank then click option'

Asolver

Bpenalty

Cmax_iter

DC

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'penalty' instead of 'C' for regularization strength.

Not prefixing the parameter with 'clf__' to target the pipeline step.

The 'C' parameter controls the inverse of regularization strength in logistic regression.

3fill in blank

hard

Fix the error in the GridSearchCV instantiation by completing the missing argument.

ML Python

from sklearn.model_selection import GridSearchCV

grid_search = GridSearchCV(pipeline, param_grid, [1]=5)

Drag options to blanks, or click blank then click option'

Acv

Bscoring

Cverbose

Dn_jobs

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'scoring' instead of 'cv' for cross-validation folds.

Omitting the 'cv' argument leading to default behavior.

The 'cv' argument sets the number of cross-validation folds in GridSearchCV.

4fill in blank

hard

Fill both blanks to fit the grid search on training data and get the best parameters.

ML Python

grid_search.[1](X_train, y_train)
best_params = grid_search.[2]

Drag options to blanks, or click blank then click option'

Afit

Bpredict

Cbest_params_

Dscore

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'predict' instead of 'fit' to train the model.

Trying to access 'best_params' without the trailing underscore.

You fit the grid search with training data and then access best_params_ to get the best parameters.

5fill in blank

hard

Fill all three blanks to predict on test data, calculate accuracy, and print the result.

ML Python

from sklearn.metrics import [1]
predictions = grid_search.[2](X_test)
accuracy = [3](y_test, predictions)
print(f'Accuracy: {accuracy:.2f}')

Drag options to blanks, or click blank then click option'

Aaccuracy_score

Bpredict

Dconfusion_matrix

Attempts:

3 left

💡 Hint

Common Mistakes

Using confusion_matrix instead of accuracy_score for metric.

Calling predict on pipeline instead of grid_search.

Import accuracy_score, use grid_search.predict to get predictions, then calculate accuracy_score with true labels.