matlab - Suppyling a hessian to fmin_ncg in python -

for scipy function fmin_ncg, there way of suppling hessian , gradient variable rather function?

i'm trying rewrite matlab code in python. code involves using optimisation routine fit parameters set of data. this, i've supplied gradient , hessian. e.g in matlab have this:

fmincon(@myfunc,x0,[],[],[],[],lb,ub,[],options); 

where myfunc returns 3 values: function evaluation, gradient, , hessian.

however fmin_ncg in python appears gradiant , hessian must supplied separate functions.

to me seems inefficient code has go through large dataset, , there calculations common function, gradient , hessian. e.g. imagine function f(x) = a(x)*b(x) gradient g(x) = a(x)*c(x), hessian h(x) = a(x)*d(x) ... in matlab can calculate a(x) once, appears have calculate 3 times in python.

have misunderstood how fmin_ncg works or there way around this?

you can create class includes of functions. each iteration, common variables calculated during first function call, reused other calls. callback feature of fmin_ncg can used reset common variables @ end of each iteration.

class function(object):     def __init__(self):       self.commonvarsdirty = true     def calcfunction(self,x,*args,**kwargs):       if self.commonvarsdirty:          self.calccommonvars()       return self.a*b     def calcgradient(self,x,*args,**kwargs):       if self.commonvarsdirty:          self.calccommonvars()       return self.a*c     def calchessian(self,x,*args,**kwargs):       if self.commonvarsdirty:          self.calccommonvars()       return self.a*d     def resetcommonvars(self,*args,**kwargs):       self.commonvarsdirty = true     def calccommonvars(self):       self.commonvarsdirty = false       # calculate common variables , save them class attributes       self.a = 1+1 

you use this.

f = function() fmin_ncg(f.calcfunction,x0,f.calcgradient,fhess=f.calchessian,callback=f.resetcommonvars) 

this adds overhead worth if computation effort calculate common variables significant.