06. Training models in sklearn
Training Models in scikit learn
In this section, we'll still be working with the dataset of the previous sections.
In the last section, we learned one of the most important classification algorithms in Machine Learning, including the following:
- Logistic Regression
- Neural Networks
- Decision Trees
- Support Vector Machines
Now, we'll have the chance to use them in real data! In sklearn, this is very easy, all we do is define our classifier, and then use the following line to fit the classifier to the data (which we call X, y):
classifier.fit(X,y)Here are the main classifiers we define, together with the package we must import:
Logistic Regression
from sklearn.linear_model import LogisticRegression
classifier = LogisticRegression()Neural Networks
(note: This is only available on versions 0.18 or higher of scikit-learn)
from sklearn.neural_network import MLPClassifier
classifier = MLPClassifier()Decision Trees
from sklearn.tree import DecisionTreeClassifier
classifier = DecisionTreeClassifier()Support Vector Machines
from sklearn.svm import SVC
classifier = SVC()Example: Logistic Regression
Let's do an end-to-end example on how to read data and train our classifier. Let's say we carry our X and y from the previous section. Then, the following commands will train the Logistic Regression classifier:
from sklearn.linear_model import LogisticRegression
classifier = LogisticRegression()
classifier.fit(X,y)This gives us the following boundary:
Quiz: Train your own model
Now, it's your turn to shine! In the quiz below, we'll work with the following dataset:
Your goal is to use one of the classifiers above, between Logistic Regression, Decision Trees, or Support Vector Machines (sorry, Neural Networks are still not available in this version of sklearn, but we will be upgrading soon!), to see which one will fit the data better. Click on Test Run to see the graphical output of your classifier, and in the quiz underneath this, enter the classifier that you think fit the data better!
Start Quiz:
Training Models Quiz
SOLUTION:
- Decision Tree
- Support Vector Machine