09.07.

2019 Sc k t-Learn Cheat Sheet: Python Mach ne Learn ng (art cle) - DataCamp

Scikit-Learn Cheat Sheet: Python Machine

Learning
A handy sc k t-learn cheat sheet to mach ne learn ng w th Python, nclud ng
code examples.

Most of you who are learn ng data sc ence w th Python w ll have def n tely heard already about
scikit-learn , the open source Python l brary that mplements a w de var ety of mach ne learn ng,
preprocess ng, cross-val dat on and v sual zat on algor thms w th the help of a un f ed nterface.

If you're st ll qu te new to the f eld, you should be aware that mach ne learn ng, and thus also th s
Python l brary, belong to the must-knows for every asp r ng data sc ent st.

That's why DataCamp has created a scikit-learn cheat sheet for those of you who have already
started learn ng about the Python package, but that st ll want a handy reference sheet. Or, f you st ll
have no dea about how scikit-learn works, th s mach ne learn ng cheat sheet m ght come n handy
to get a qu ck f rst dea of the bas cs that you need to know to get started.

E ther way, we're sure that you're go ng to f nd t useful when you're tackl ng mach ne learn ng
problems!

Th s scikit-learn cheat sheet w ll ntroduce you to the bas c steps that you need to go through to
mplement mach ne learn ng algor thms successfully: you'll see how to load n your data, how to
preprocess t, how to create your own model to wh ch you can f t your data and pred ct target labels,
how to val date your model and how to tune t further to mprove ts performance.

https://www.datacamp.com/commun ty/blog/sc k t-learn-cheat-sheet 1/9

09.07.2019 Sc k t-Learn Cheat Sheet: Python Mach ne Learn ng (art cle) - DataCamp

In short, th s cheat sheet w ll k ckstart your data sc ence projects: w th the help of code examples, you'll
have created, val dated and tuned your mach ne learn ng models n no t me.

So what are you wa t ng for? T me to get started!

(Cl ck above to download a pr ntable vers on or read the onl ne vers on below.)

Python For Data Science Cheat Sheet: Scikit-learn

Sc k t-learn s an open source Python l brary that mplements a range of mach ne learn ng,
preprocess ng, cross-val dat on and v sual zat on algor thms us ng a un f ed nterface.

A Basic Example

>>> from sklearn import neighbors, datasets, preprocessing

>>> from sklearn.model_selection import train_test_split
>>> from sklearn.metrics import accuracy_score
https://www.datacamp.com/commun ty/blog/sc k t-learn-cheat-sheet 2/9
09.07.2019 Sc k t-Learn Cheat Sheet: Python Mach ne Learn ng (art cle) - DataCamp

>>> iris = datasets.load_iris()

>>> X, y = iris.data[:, :2], iris.target
>>> X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=33)
>>> scaler = preprocessing.StandardScaler().fit(X_train)
>>> X_train = scaler.transform(X_train)
>>> X_test = scaler.transform(X_test)
>>> knn = neighbors.KNeighborsClassifier(n_neighbors=5)
>>> knn.fit(X_train, y_train)
>>> y_pred = knn.predict(X_test)
>>> accuracy_score(y_test, y_pred)

Loading The Data

Your data needs to be numer c and stored as NumPy arrays or Sc Py sparse matr ces. Other types that
are convert ble to numer c arrays, such as Pandas DataFrame, are also acceptable.

>>> import numpy as np

>>> X = np.random.random((10,5))
>>> y = np.array(['M','M','F','F','M','F','M','M','F','F','F'])
>>> X[X < 0.7] = 0

Preprocessing The Data

Standardization

>>> from sklearn.preprocessing import StandardScaler

>>> scaler = StandardScaler().fit(X_train)
>>> standardized_X = scaler.transform(X_train)
>>> standardized_X_test = scaler.transform(X_test)

Normalization

>>> from sklearn.preprocessing import Normalizer

>>> scaler = Normalizer().fit(X_train)
>>> normalized_X = scaler.transform(X_train)
>>> normalized_X_test = scaler.transform(X_test)

Binarization

https://www.datacamp.com/commun ty/blog/sc k t-learn-cheat-sheet 3/9

09.07.2019 Sc k t-Learn Cheat Sheet: Python Mach ne Learn ng (art cle) - DataCamp

>>> from sklearn.preprocessing import Binarizer

>>> binarizer = Binarizer(threshold=0.0).fit(X)
>>> binary_X = binarizer.transform(X)

Encoding Categorical Features

>>> from sklearn.preprocessing import LabelEncoder

>>> enc = LabelEncoder()
>>> y = enc.fit_transform(y)

Imputing Missing Values

>>>from sklearn.preprocessing import Imputer

>>>imp = Imputer(missing_values=0, strategy='mean', axis=0)
>>>imp.fit_transform(X_train)

Generating Polynomial Features

>>> from sklearn.preprocessing import PolynomialFeatures

>>> poly = PolynomialFeatures(5)
>>> oly.fit_transform(X)

Training And Test Data

>>> from sklearn.model_selection import train_test_split

>>> X_train, X_test, y_train, y_test = train_test_split(X,y,random_state=0)

Create Your Model

Supervised Learning Estimators

L near Regress on

>>> from sklearn.linear_model import LinearRegression

>>> lr = LinearRegression(normalize=True)

Support Vector Mach nes (SVM)

https://www.datacamp.com/commun ty/blog/sc k t-learn-cheat-sheet 4/9

09.07.2019 Sc k t-Learn Cheat Sheet: Python Mach ne Learn ng (art cle) - DataCamp

>>> from sklearn.svm import SVC

>>> svc = SVC(kernel='linear')

Na ve Bayes

>>> from sklearn.naive_bayes import GaussianNB

>>> gnb = GaussianNB()

KNN

>>> from sklearn import neighbors

>>> knn = neighbors.KNeighborsClassifier(n_neighbors=5)

Unsuperv sed Learn ng Est mators

Pr nc pal Component Analys s (PCA)

>>> from sklearn.decomposition import PCA

>>> pca = PCA(n_components=0.95)

K Means

>>> from sklearn.cluster import KMeans

>>> k_means = KMeans(n_clusters=3, random_state=0)

Model Fitting

Supervised learning

>>> lr.fit(X, y)
>>> knn.fit(X_train, y_train)
>>> svc.fit(X_train, y_train)

Unsuperv sed Learn ng

>>> k_means.fit(X_train)
fi f (
https://www.datacamp.com/commun ty/blog/sc k t-learn-cheat-sheet
i ) 5/9
09.07.2019 Sc k t-Learn Cheat Sheet: Python Mach ne Learn ng (art cle) - DataCamp
>>> pca_model = pca.fit_transform(X_train)

Prediction

Superv sed Est mators

>>> y_pred = svc.predict(np.random.random((2,5)))

>>> y_pred = lr.predict(X_test)

>>> y_pred = knn.predict_proba(X_test))

Unsuperv sed Est mators

>>> y_pred = k_means.predict(X_test)

Evaluate Your Model's Performance

Classification Metrics

Accuracy Score

>>> knn.score(X_test, y_test)

>>> from sklearn.metrics import accuracy_score
>>> accuracy_score(y_test, y_pred)

Class f cat on Report

>>> from sklearn.metrics import classification_report

>>> print(classification_report(y_test, y_pred)))

Confus on Matr x

>>> from sklearn.metrics import confusion_matrix

>>> print(confusion_matrix(y_test, y_pred)))

Regression Metrics
https://www.datacamp.com/commun ty/blog/sc k t-learn-cheat-sheet 6/9
09.07.2019 Sc k t-Learn Cheat Sheet: Python Mach ne Learn ng (art cle) - DataCamp

Mean Absolute Error

>>> from sklearn.metrics import mean_absolute_error

>>> y_true = [3, -0.5, 2])
>>> mean_absolute_error(y_true, y_pred))

Mean Squared Error

>>> from sklearn.metrics import mean_squared_error

>>> mean_squared_error(y_test, y_pred))

R2 Score

>>> from sklearn.metrics import r2_score

>>> r2_score(y_true, y_pred))

Clustering Metrics

Adjusted Rand Index

>>> from sklearn.metrics import adjusted_rand_score

>>> adjusted_rand_score(y_true, y_pred))

Homogene ty

>>> from sklearn.metrics import homogeneity_score

>>> homogeneity_score(y_true, y_pred))

V-measure

>>> from sklearn.metrics import v_measure_score

>>> metrics.v_measure_score(y_true, y_pred))

Cross-Validation

>>> print(cross_val_score(knn, X_train, y_train, cv=4))

https://www.datacamp.com/commun ty/blog/sc k t-learn-cheat-sheet 7/9

09.07.2019 Sc k t-Learn Cheat Sheet: Python Mach ne Learn ng (art cle) - DataCamp

>>> print(cross_val_score(lr, X, y, cv=2))

Tune Your Model

Grid Search

>>> from sklearn.grid_search import GridSearchCV

>>> params = {"n_neighbors": np.arange(1,3), "metric": ["euclidean", "cityblock"]}

>>> grid = GridSearchCV(estimator=knn,param_grid=params)

>>> grid.fit(X_train, y_train)

>>> print(grid.best_score_)

>>> print(grid.best_estimator_.n_neighbors)

Randomized Parameter Optimization

>>> from sklearn.grid_search import RandomizedSearchCV

>>> params = {"n_neighbors": range(1,5), "weights": ["uniform", "distance"]}

>>> rsearch = RandomizedSearchCV(estimator=knn,

param_distributions=params,
cv=4,
n_iter=8,
random_state=5)

>>> rsearch.fit(X_train, y_train)

>>> print(rsearch.best_score_)

https://www.datacamp.com/commun ty/blog/sc k t-learn-cheat-sheet 8/9

09.07.2019 Sc k t-Learn Cheat Sheet: Python Mach ne Learn ng (art cle) - DataCamp

Going Further
Beg n w th our sc k t-learn tutor al for beg nners, n wh ch you'll learn n an easy, step-by-step way how
to explore handwr tten d g ts data, how to create a model for t, how to f t your data to your model and
how to pred ct target values. In add t on, you'll make use of Python's data v sual zat on l brary
matplotl b to v sual ze your results.

https://www.datacamp.com/commun ty/blog/sc k t-learn-cheat-sheet 9/9