Linear svm

import numpy as np

class SimpleSVM:
    def __init__(self, learning_rate=0.001, lambda_param=0.01, n_iters=1000):
        self.lr = learning_rate
        self.lambda_param = lambda_param
        self.n_iters = n_iters
        self.weights = None
        self.bias = None

    def fit(self, X, y):
        n_samples, n_features = X.shape
        y_ = np.where(y <= 0, -1, 1)

        self.weights = np.zeros(n_features)
        self.bias = 0
        """
        During training, we iterate through each sample in the dataset. 
        The condition checks if the sample is on the correct side of the margin. 
        If it is, we only update the weights based on the regularization term. 
        If it's not, we update the weights and bias based on both the regularization term and the misclassification.
        """
        for _ in range(self.n_iters):
            for idx, x_i in enumerate(X):
                condition = y_[idx] * (np.dot(x_i, self.weights) - self.bias) >= 1
                if condition:
                    self.weights -= self.lr * (2 * self.lambda_param * self.weights)
                else:
                    self.weights -= self.lr * (2 * self.lambda_param * self.weights - np.dot(x_i, y_[idx]))
                    self.bias -= self.lr * y_[idx]

    def predict(self, X):
        linear_output = np.dot(X, self.weights) - self.bias
        return np.sign(linear_output)

---

Last update: March 21, 2024