Search⌘ K
AI Features
Home
Courses
Hands-on Machine Learning with Scikit-Learn

Challenge Solution Review

Explore how to solve and evaluate a neural network classification challenge using Scikit-Learn's MLPClassifier. Learn to train models, adjust parameters, and assess predictions with F1-score for better understanding of deep learning applications.

We'll cover the following...
Python 3.5
import sklearn.datasets as datasets
from sklearn.neural_network import MLPClassifier
from sklearn.model_selection import train_test_split
import sklearn.metrics as metrics
X, y = datasets.make_classification(n_samples=1000,
                                    n_features=30,
                                    random_state=10)
train_x, test_x, train_y, test_y = train_test_split(X,
                                                    y,
                                                    test_size=0.2,
                                                    random_state=42)
nn = MLPClassifier(batch_size=32,
                   hidden_layer_sizes=(64, 32),
                   solver="sgd",
                   shuffle=True,
                   tol=1e-3,
                   max_iter=500,
                   learning_rate_init=0.0001, random_state=13)
nn.fit(train_x, train_y)
pred_y = nn.predict(test_x)
f1 = metrics.f1_score(y_true=test_y, y_pred=pred_y)
print("The F1 score is {}.".format(f1))
...