Preparing data
- Machine Learning course
- Beginner
- Lesson 93/112
- 7 min
Prepare data for regression.
Project: disease progression
In this project you'll predict the disease progression one year later (a numeric score) for 442 diabetes patients from age, sex, body mass index, blood pressure and 6 blood measurements. This is a regression problem.
from sklearn.datasets import load_diabetes
d = load_diabetes()
print(d.data.shape) # (442, 10)
print(d.feature_names) # ['age', 'sex', 'bmi', 'bp', 's1', ...]
print(d.target[:5]) # [151. 75. 141. 206. 135.]The features in this dataset are pre-scaled: each column's mean is close to 0.
Getting to know the target
In a regression project, the target's range and distribution are needed to interpret the size of the error: an MAE of 50 means one thing if the target is in 25-350 and another if it is in 0-60.
print(y.min(), y.max()) # 25.0 346.0
print(round(y.mean(), 1), round(y.std(), 1))As a baseline, the MAE of the "predict the mean for every patient" model is computed; the model must beat it.
Practice Training and test sets
Split the data with test_size=0.2, random_state=0. Print the training and test sizes on one line, then the mean of the training target with round(..., 1).
Starter code
from sklearn.datasets import load_diabetes
from sklearn.model_selection import train_test_split
X, y = load_diabetes(return_X_y=True)
# buraya yazSample solution
from sklearn.datasets import load_diabetes
from sklearn.model_selection import train_test_split
X, y = load_diabetes(return_X_y=True)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=0)
print(len(X_train), len(X_test))
print(round(y_train.mean(), 1))Expected output
353 89 151.6
Questions and answers
Review the lesson questions together with their answers. To test yourself, solve the lesson interactively.
1 Which family of metrics should be used in this project?
- Accuracy, F1
- Confusion matrix
- Precision, recall
- MAE, RMSE, R² Correct answer
Explanation: The target is a continuous score, so regression metrics are used.
2 Set aside 20% of the data for testing.
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=____, random_state=0)X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=0)Explanation: stratify isn't used in regression; there are no classes.
3 Match the facts.
| Term | Meaning |
|---|---|
442 |
Number of patients |
10 |
Number of features |
bmi |
Body mass index |
target |
One-year progression |