Source code for snowdrop.src.numeric.estimation.test_mle

import numpy as np

[docs] def test(): from mle import var,Normal # Define model x = var('x', observed=True, vector=True) y = var('y', observed=True, vector=True) a = var('a') b = var('b') sigma = var('sigma') model = Normal(y, a * x + b, sigma) # Generate data xs = np.linspace(0, 2, 20) ys = 0.5 * xs + 0.3 + np.random.normal(0, 0.1, 20) # Fit model to data result = model.fit({'x': xs, 'y': ys}, {'a': 1, 'b': 1, 'sigma': 1}) print(result)
if __name__ == '__main__': """Test program.""" test()