Practice Train/test Split (3) - Classification Algorithms - Data Science Basic

Practice: Train/Test Split

Lesson 3Classification Algorithms

Practise this section

9 questions

Flashcard drill
3 cards from this section. Quick, good the night before a test.
Exercise
6 questions · about 4 min · pass mark 70%. Hints available.
Sign up to start the exercise
Quiz
3 questions · about 2 min · pass mark 70%. Hints available.
Sign up to start the quiz

6 sample questions

Question 1.
What is the purpose of the Train/Test Split?
Show hintHide hint
Think about why testing on unseen data is necessary.
Question 2.
What can you infer if your model performs well on the training set but poorly on the test set?
Show hintHide hint
Consider model bias and prediction generalization.
3Question 3.
What is the primary goal of a Train/Test Split?
- A) To increase the dataset size
- B) To create a validation set
- C) To evaluate model performance on unseen data
- D) To reduce the training time
Show hintHide hint
Consider what the primary function of separating data into two parts is.
4Question 4.
True or False: The 'random_state' parameter ensures reproducibility of the Train/Test Split.
- True
- False
Show hintHide hint
Think about why reproducing results are important in data science.
Question 5.
Consider a dataset of 10,000 samples. If you want to perform an 80/20 train/test split, how many samples would be used for training and testing?
Show hintHide hint
Calculate 20% of the total samples for testing.
Question 6.
In a scenario where your model performs excellently on the training data yet poorly on test data, what potential issues could cause this discrepancy? Discuss and suggest solutions.
Show hintHide hint
Think about the relationship between training data, model complexity, and generalization.

5 more questions in this section. Enroll free to practise them all.