data.py

import numpy as np

class Dataset:
    def __init__(self, means1=[-5, 0], means2=[0, 5]):
        self._rand_sample(means1, means2)

    def _rand_sample(self, means1, means2):
        means1 = np.array(means1)
        means2 = np.array(means2)
        covar = np.array([1, 0, 0, 1]).reshape(2,2)
        x1 = np.random.multivariate_normal(means1, covar, size=200)
        x2 = np.random.multivariate_normal(means2, covar, size=200)
        y1 = np.ones((200, 1))
        y2 = np.ones((200, 1)) * -1
        #y2 = np.ones((200, 1)) * -1
        self._split(x1, y1, x2, y2)

    def _split(self, x1, y1, x2, y2):
        num1 = x1.shape[0]
        train_num1 = int(num1 * 0.8)
        num2 = x2.shape[0]
        train_num2 = int(num2 * 0.8)

        self.x_train = np.concatenate((x1[:train_num1],x2[:train_num2]),axis=0)
        self.y_train = np.concatenate((y1[:train_num1],y2[:train_num2]),axis=0)
        self.x_test = np.concatenate((x1[train_num1:],x2[train_num2:]),axis=0)
        self.y_test = np.concatenate((y1[train_num1:],y2[train_num2:]),axis=0)