Python Basics with Numpy

Numpy를 이용해 기본적인 함수 만들기

Sigmoid function

def sigmoid(x):
    s = 1/(1+np.exp(-x))
    return s
    
t_x = np.array([1,2,3])
print(sigmoid(t_x))

Sigmoid Gradient

sigmoid_derivative(x)=f(x)(1-f(x))

def sigmoid_derivative(x):
    s=sigmoid(x)
    ds=s*(1-s)
    return ds
t_x = np.array([1,2,3])
sigmoid_derivative(t_x)

Reshaping arrays

(length, height, 3)를 넣으면 벡터값(length x height x3, 1)을 리턴해주는 함수

def image2vector(image):
    v = image.reshape((image.shape[0]*image.shape[1]*image.shape[2],1))
    return v

Normalizing rows

def normalize_rows(x):
    x_norm = np.linalg.norm(x, axis=1, keepdims = True)
    x = x / x_norm

axis = 1, keepsdim =1 에 대한 설명은 여기 참고하면 도움이 된다.

Softmax(다중 클래스 분류)

def softmax(x):
    x_exp = np.exp(x)
    x_sum = np.sum(x_exp, axis=1, keepdims=True)
    s = x_exp / x_sum
    return s

Loss functions

def L1(yhat, y):
    loss = np.sum(np.abs(yhat-y) , axis = 0)
    return loss

def L2(yhat, y):
    loss = np.dot(np.abs(yhat-y), np.abs(yhat-y))
    return loss