Custom Loss Function for Linear Regression in Python
from scipy.optimize import minimize
def error(y_true, y_pred):
...
def loss(params, x, y_true):
pred = ...params...
return ...loss(y_true, pred)...
d = minimize(loss, [0,1], args=(x, y_true))
res = d.x