Pipeline with GridSearchCV in ML Python - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of GridSearchCV best parameters in pipeline

What will be the output of the following code snippet after fitting the GridSearchCV pipeline?

ML Python

from sklearn.datasets import load_iris
from sklearn.model_selection import GridSearchCV
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler
from sklearn.svm import SVC

iris = load_iris()
X, y = iris.data, iris.target

pipe = Pipeline([
    ('scaler', StandardScaler()),
    ('svc', SVC())
])

param_grid = {
    'svc__C': [0.1, 1],
    'svc__kernel': ['linear', 'rbf']
}

gs = GridSearchCV(pipe, param_grid, cv=3)
gs.fit(X, y)

print(gs.best_params_)

A {'svc__C': 0.1, 'svc__kernel': 'rbf'}

B {'svc__C': 0.1, 'svc__kernel': 'linear'}

C {'svc__C': 1, 'svc__kernel': 'linear'}

D {'svc__C': 1, 'svc__kernel': 'rbf'}

💡 Hint

Think about which kernel usually performs better on iris dataset with default settings.

The RBF kernel with C=1 generally performs better on the iris dataset than linear kernel or smaller C values, so GridSearchCV selects {'svc__C': 1, 'svc__kernel': 'rbf'}.

❓ Model Choice

intermediate

Choosing the correct pipeline step for text data

You want to build a pipeline for text classification using GridSearchCV. Which pipeline step should you include before the classifier to convert text into numbers?

A CountVectorizer()

B StandardScaler()

C PCA()

D MinMaxScaler()

💡 Hint

Text data needs to be converted into numeric features before classification.

CountVectorizer converts text into a matrix of token counts, which is suitable for text classification pipelines.

❓ Hyperparameter

advanced

Correct hyperparameter name in pipeline for GridSearchCV

Given a pipeline named 'pipe' with steps [('scaler', StandardScaler()), ('clf', RandomForestClassifier())], which is the correct hyperparameter name to tune the number of trees in GridSearchCV?

A 'n_estimators'

B 'pipe__n_estimators'

C 'clf__n_estimators'

D 'scaler__n_estimators'

💡 Hint

In pipelines, hyperparameters are prefixed by the step name and two underscores.

To tune hyperparameters of a step in a pipeline, use 'stepname__parameter'. Here, 'clf' is the step name, so 'clf__n_estimators' is correct.

❓ Metrics

advanced

Evaluating GridSearchCV best score attribute

After fitting GridSearchCV with cv=5, what does the attribute 'best_score_' represent?

A The training score of the best estimator on the full training data

B The mean cross-validation score of the best estimator across the 5 folds

C The test score on unseen data

D The highest score from a single fold during cross-validation

💡 Hint

GridSearchCV uses cross-validation to estimate performance.

'best_score_' is the average cross-validation score for the best parameter set found during GridSearchCV.

🔧 Debug

expert

Identifying error in pipeline parameter grid

What error will the following GridSearchCV code raise? from sklearn.pipeline import Pipeline from sklearn.preprocessing import StandardScaler from sklearn.linear_model import LogisticRegression from sklearn.model_selection import GridSearchCV pipe = Pipeline([ ('scaler', StandardScaler()), ('logreg', LogisticRegression()) ]) param_grid = { 'C': [0.1, 1, 10] } gs = GridSearchCV(pipe, param_grid) gs.fit(X, y)

A KeyError because 'C' is not a valid parameter name for the pipeline

B TypeError because LogisticRegression() is missing required arguments

C ValueError because param_grid is empty

D No error, code runs successfully

💡 Hint

In pipelines, hyperparameters must be prefixed by the step name.

The param_grid key 'C' is invalid because it lacks the step prefix 'logreg__'. This causes a KeyError during GridSearchCV.