def get_model(
model, xdata, ydata
): # todo classfactory, and xdata y data optional (is already classfactory)
if model == 'exp_decay':
a = Parameter(name='a', value=ydata[0], min=0)
b = Parameter(name='b', value=ydata[-1], min=0)
tau_pos = np.abs(ydata - (ydata[0] / np.e + ydata[-1])).argmin()
tau = Parameter(name='tau', value=xdata[tau_pos], min=0)
x = Variable(name='x')
m = a * exp(-x / tau) + b
return m
elif model == 'double_exp_decay':
a1 = Parameter(name='a1', value=ydata[0] / 2.1, min=0)
a2 = Parameter(name='a2', value=ydata[0] / 1.9, min=0)
b = Parameter(name='b', value=ydata[-1], min=0)
tau_pos = np.abs(ydata - (ydata[0] / np.e + ydata[-1])).argmin()
tau1 = Parameter(name='tau1', value=0.5 * xdata[tau_pos], min=0)
tau2 = Parameter(name='tau2', value=2 * xdata[tau_pos], min=0)
x = Variable(name='x')
m = a1 * exp(-x / tau1) + a2 * exp(-x / tau2) + b
return m
def test_likelihood_fitting(self):
"""
Fit using the likelihood method.
"""
b = Parameter(4, min=3.0)
x = Variable()
pdf = (1/b) * exp(- x / b)
# Draw 100 points from an exponential distribution.
# np.random.seed(100)
xdata = np.random.exponential(5, 100000)
fit = Likelihood(pdf, xdata)
fit_result = fit.execute()
self.assertAlmostEqual(fit_result.params.b, 5., 1)
from __future__ import print_function from symfit.api import Parameter, Variable, Fit, exp import numpy as np import matplotlib.pyplot as plt import seaborn as sns palette = sns.color_palette() x = Variable() A = Parameter() sig = Parameter(name='sig', value=1.4, min=1.0, max=2.0) x0 = Parameter(name='x0', value=15.0, min=0.0) # Gaussian distrubution model = A*exp(-((x - x0)**2/(2 * sig**2))) # Sample 10000 points from a N(15.0, 1.5) distrubution sample = np.random.normal(loc=15.0, scale=1.5, size=(10000,)) ydata, bin_edges = np.histogram(sample, 100) xdata = (bin_edges[1:] + bin_edges[:-1])/2 fit = Fit(model, xdata, ydata) fit_result = fit.execute() print(fit_result) print(model) y = model(x=xdata, **fit_result.params) sns.regplot(xdata, ydata, fit_reg=False) plt.plot(xdata, y, color=palette[2]) plt.ylim(0, 400) plt.show()
def boltzmann(T, V, *args, **kwargs):
kbT = kb * T
return exp(-V(*args, **kwargs) / kbT)