cross entropy is correct?

[liyunrui]

No description provided.

[liyunrui]

class CrossEntropy(Loss):
def init(self): pass

def loss(self, y, p):
    # Avoid division by zero
    p = np.clip(p, 1e-15, 1 - 1e-15)
    return - y * np.log(p) - (1 - y) * np.log(1 - p)

def acc(self, y, p):
    return accuracy_score(np.argmax(y, axis=1), np.argmax(p, axis=1))

def gradient(self, y, p):
    # Avoid division by zero
    p = np.clip(p, 1e-15, 1 - 1e-15)
    return - (y / p) + (1 - y) / (1 - p)

loss = CrossEntropy()
y = np.array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])
p = np.random.uniform(0,1,size=10)
loss.loss(y, p)

[zpengc]

def acc(self, y, p):
return accuracy_score(y, p)
is this ok?