如何在python中实现负二项式损失函数以用于轻型GBM?
我有一个机器学习的问题,我相信负二项损失函数会很好地适合,但是light gbm软件包没有它作为标准,我正在尝试实现它,但我不知道如何获得梯度和Hessian,有人知道我如何做到这一点吗?我设法得到了损失函数,但我不能得到梯度和hessian如何在python中实现负二项式损失函数以用于轻型GBM?,python,machine-learning,gradient,lightgbm,hessian-matrix,Python,Machine Learning,Gradient,Lightgbm,Hessian Matrix,我有一个机器学习的问题,我相信负二项损失函数会很好地适合,但是light gbm软件包没有它作为标准,我正在尝试实现它,但我不知道如何获得梯度和Hessian,有人知道我如何做到这一点吗?我设法得到了损失函数,但我不能得到梯度和hessian import math def custom_asymmetric_valid(y_pred,y_true): y_true = y_true.get_label() p = 0.5 n = y_pred loss = m
import math
def custom_asymmetric_valid(y_pred,y_true):
y_true = y_true.get_label()
p = 0.5
n = y_pred
loss = math.gamma(n) + math.gamma(y_true + 1) - math.gamma(n + y_true) - n * math.log(p) - y_true * math.log(1 - p)
return "custom_asymmetric_eval", np.mean(loss), False
现在如何得到梯度和Hessian
def custom_asymmetric_train(y_pred,y_true):
residual = (y_true.get_label() - y_pred).astype("float")
grad = ?
hess = ?
return grad, hess
任何人都可以提供帮助?这可以通过scipy自动实现:
from scipy.misc import derivative
from scipy.special import gamma
def custom_asymmetric_train(y_pred, dtrain):
y_true = dtrain.label
p = 0.5
def loss(x,t):
loss = gamma(x) + gamma(t+1) - gamma(x+t) - x*np.log(p) - t*np.log(1-p)
return loss
partial_d = lambda x: loss(x, y_true)
grad = derivative(partial_d, y_pred, n=1, dx=1e-6)
hess = derivative(partial_d, y_pred, n=2, dx=1e-6)
return grad, hess