Challenge: Evaluating the Perceptron
To evaluate the previously created perceptron, you will use a dataset containing two input features and two distinct classes (0 and 1):
This dataset is balanced, with 500 samples from class 1 and 500 samples from class 0. Therefore, accuracy is a sufficient metric for evaluation in this case, which can be calculated using the accuracy_score() function:
accuracy_score(y_true, y_pred)
y_true represents the actual labels, while y_pred represents the predicted labels.
The dataset is stored in perceptron.py as two NumPy arrays: X (input features) and y (corresponding labels), so they will be simply imported. This file also contains model, which is the instance of the Perceptron class you previously created.
Swipe to start coding
Obtain predictions from the trained model and evaluate its performance:
- Split the dataset into training (80%) and testing (20%) sets.
- Train the model for 10 epochs with a learning rate of
0.01. - Obtain predictions for all examples in the test set.
- Calculate the accuracy by comparing the predicted labels with the actual test labels.
Lösning
Byt till skrivbordet för praktisk övningFortsätt där du är med ett av alternativen nedanTack för dina kommentarer!
