def __call__(self): if self.model!='nugget': # what are 3 and 10? should these be parametrizable? params = fit_function(self.x, self.y, self.model, 3, 10) else: # what are 1 and 1? should these be parametrizable? params = fit_function(self.x, self.y, self.model, 1, 1) return self.model(self.x, params)
def __call__(self): if self.model_text != 'nugget': # what are 3 and 10? should these be parametrizable? params = fit_function(self.x, self.y, self.model, 3, 10) y = self.model(self.x, params) else: # what are 1 and 1? should these be parametrizable? params = fit_function(self.x, self.y, self.model, 1, 1) y = [self.model(self.x, params)] * len(self.x) return y, params, self.text
def __call__(self): if self.model_text!='nugget': # what are 3 and 10? should these be parametrizable? params = fit_function(self.x, self.y, self.model, 3, 10) y = self.model(self.x, params) else: # what are 1 and 1? should these be parametrizable? params = fit_function(self.x, self.y, self.model, 1, 1) y = [self.model(self.x, params)] * len(self.x) return y, params, self.text
def test_exponential(self): """test exponential approximation""" # defining our fitting function def f(x,a): return exp(a[0]+x*a[1]) exp_params = [2, 10] x = arange(-1,1,.01) y = f(x, exp_params) y_noise = y + rand(len(y)) params = fit_function(x, y_noise, f, 2, 5) self.assertFloatEqual(params, exp_params , .5)
def test_constant(self): """test constant approximation""" # defining our fitting function def f(x,a): return a[0] exp_params = [2] x = arange(-1,1,.01) y = f(x, exp_params) y_noise = y + rand(len(x)) params = fit_function(x, y_noise, f, 1, 5) self.assertFloatEqual(params, exp_params , .5)