AI Linear regression 예제

개념 내용은 이전 글 참조바랍니당

2020/05/24 - [AI] - AI Linear regression 파이썬 구현

AI Linear regression 파이썬 구현

 

입력 데이터

sps.csv

0.00MB

 

import numpy as np

try:
    load_data = np.loadtxt('./sps.csv', delimiter=',', dtype=np.float32)

    y_data = load_data[:, [0]]
    x_data = load_data[:, 1:]

    W = np.random.rand(4, 1)
    b = np.random.rand(1)


    def loss_func(x, t):
        y = np.dot(x, W) + b
        return (np.sum((t - y) ** 2)) / len(x)


    def predict(x):
        return np.dot(x, W) + b


    def numerical_derivative(f, input_data):
        delta_x = 1e-4
        ret = np.zeros_like(input_data)

        it = np.nditer(input_data, flags=['multi_index'], op_flags=['readwrite'])

        while not it.finished:
            idx = it.multi_index

            tmp = input_data[idx]
            input_data[idx] = float(tmp) - delta_x
            fx1 = f(input_data)

            input_data[idx] = float(tmp) + delta_x
            fx2 = f(input_data)

            ret[idx] = (fx2 - fx1) / (2 * delta_x)

            input_data[idx] = tmp
            it.iternext()

        return ret


    learning_rate = 1e-4
    f = lambda x: loss_func(x_data, y_data)

    print('Initial loss = ', loss_func(x_data, y_data), 'W = ', W, 'b = ', b)

    for step in range(8001):
        W -= learning_rate * numerical_derivative(f, W)
        b -= learning_rate * numerical_derivative(f, b)

        if step % 400 == 0:
            print('loss = ', loss_func(x_data, y_data), 'W = ', W, 'b = ', b)

    print(predict([2, -3, 9, 0]))

except Exception as err:
    print(str(err))

 

결과