Model Training

Multiple Choice

What's the first step of model training?

• Hyperparameter tuning
• Loading the data

• Model evaluation and testing
• Algorithm selection and implementation

Drag and Drop

Put in correct order

​

Placeholder

​

Placeholder

​

Placeholder

​

Placeholder

​

Drag these tiles and drop them in the correct blank above

import

pandas

as

pd

Drag and Drop

Put in correct order

​

Placeholder

​

Placeholder

​

Placeholder

​ ​

Drag these tiles and drop them in the correct blank above

houses =

pd.read_csv

("houses.csv")

(houses.csv)

Multiple Choice

What's the purpose of data exploration?

• To visualize data in a graphical format for presentations.
• To clean the data by removing all outliers and errors.
• To store data in a database for future retrieval and analysis.
• To understand the data's structure, patterns and problems.

Multiple Choice

Why do we need to encode categorical data?

• We need to encode categorical data becayse machine learning needs numbers.

• We need to encode categorical data to simplify the data entry process.
• We need to encode categorical data to enhance the quality of textual data.
• We need to encode categorical data to improve data visualization techniques.

Multiple Choice

Which of the following columns in the houses dataset are categorical variables?

• bedrooms, area_m2

• district, apt_floor

• year_built, total_floors

• price, bathrooms

Multiple Choice

Which step in the model training process involves converting categorical variables into numerical format?

• Load the data

• Data exploration

• Data encoding

• Train the model

Multiple Choice

What do we import from sklearn to divide our data?

• from sklearn.model_selection import train_test_split
• from sklearn.preprocessing import train_test
• from sklearn.model_selection import data_split
• from sklearn.data import split_data

Multiple Choice

What's in variable y?

• Prediction

• Target

• Zero
• Features

Multiple Choice

What's in variable X?

• Prediction

• Target

• Zero
• Features

Multiple Select

What doest test_size = 0.2 mean?

• test_size = 0.2 means 2% of the data is used for validation.
• test_size = 0.2 indicates 20% of the data is ignored during training.
• test_size = 0.2 means that 80% of the data is for training purposes.

• test_size = 0.2 means 20% of the data is for testing.

Multiple Choice

How to train the model?

• model.train()

• model.fit()

• model.predict()

• model.AI()

Multiple Choice

Which variables do we train the model on?

• X_test, y_test

• X_train, y_train

• X, y

• Train, test

Multiple Choice

After training the model, which step should be performed next according to the model training steps?

• Make predictions

• Evaluate the model

• Load the data

• Data encoding

Multiple Choice

How do we make predictions?

• model.predict()

• model.pred()

• model.fit()

• model.answer()

Multiple Choice

Which variable do we use for prediction?

• y_train

• y_test

• X_train

• X_test

Multiple Choice

Why do we use X_test for predictions?

• We use X_test to visualize the training process.
• We use X_test to train the model on existing data.
• We use X_test to improve the model's accuracy on training data.
• We use X_test for predictions to see model performance on new, unseen data.

Multiple Choice

Which Python function is used to make predictions on the test set in a machine learning model?

• model.fit(X_test)

• model.evaluate(X_test)

• model.predict(X_test)

• model.score(X_test)

It means: On average, our prediction is wrong by 191万.

Multiple Choice

Why do we use both y_test and y_pred when evaluating the model?

• y_pred is used to compare predictions from different models.
• We use y_test to adjust the model's parameters during training.
• y_test helps in selecting the best features for the model.
• We use y_test to compare the model's predictions (y_pred) to real data.

Multiple Choice

The bigger $$MAE$$ the better the model.

• True

• False

Multiple Select

What's the meaning of this read line?

• predicted price = actual price

• x = y

• the model is good

• the model is bad