def test_laplace_fit():
"""Test if the fitted parameters match the true parameters."""
mu = np.array([-1, 0., 1.])
beta = np.array([0.5, 1., 2.])
nSamples = 100000
nParameters = mu.shape[0]
data = np.random.random((nSamples, nParameters))
norm = sp.stats.laplace(loc=mu, scale=beta)
data = norm.rvs(size=(nSamples, nParameters))
parameters = {"mu": tf.constant(np.ones(nParameters)),
"beta": tf.constant(np.ones(nParameters))}
tfData = tf.constant(data)
parameters = LaplaceAlgorithms.fit(parameters=parameters,
data=tfData)
with tf.Session() as sess:
parameters = sess.run(parameters)
muHat = parameters["mu"]
assert(muHat.shape == mu.shape)
assert(np.allclose(muHat, mu, atol=1e-1))
betaHat = parameters["beta"]
assert(betaHat.shape == beta.shape)
assert(np.allclose(betaHat, beta, atol=1e-1))
def test_laplace_mode():
"""Test if the mode is equal to `mu`."""
mu = np.array([-1, 0., 1.])
beta = np.array([0.5, 1., 2.])
nParameters = mu.shape[0]
parameters = {"mu": tf.constant(mu),
"beta": tf.constant(beta)}
mode = LaplaceAlgorithms.mode(parameters=parameters)
with tf.Session() as sess:
mode = sess.run(mode)
assert(mode.shape == (nParameters,))
assert(np.all(mode == mu))
def test_laplace_llh():
"""Test if the llh is the same as reported by scipy."""
mu = np.array([-1, 0., 1.])
beta = np.array([0.5, 1., 2.])
nSamples = 1000
nParameters = mu.shape[0]
data = np.random.random((nSamples, nParameters))
parameters = {"mu": tf.constant(mu),
"beta": tf.constant(beta)}
tfData = tf.constant(data)
llh = LaplaceAlgorithms.llh(parameters=parameters,
data=tfData)
with tf.Session() as sess:
llh = sess.run(llh)
assert(llh.shape == (nSamples, nParameters))
spLlh = sp.stats.laplace(loc=mu, scale=beta).logpdf(data)
assert(np.allclose(llh, spLlh))
def test_laplace_sample():
"""Test whether the mean and the variance of the samples are correct."""
mu = np.array([-1, 0., 1.])
beta = np.array([0.5, 1., 2.])
nSamples = 1000000
nParameters = mu.shape[0]
parameters = {"mu": tf.constant(mu),
"beta": tf.constant(beta)}
tfNSamples = tf.constant(nSamples)
r = LaplaceAlgorithms.sample(parameters=parameters,
nSamples=tfNSamples)
with tf.Session() as sess:
r = sess.run(r)
assert(r.shape == (nSamples, nParameters))
muHat = np.median(r, axis=0)
assert(np.allclose(muHat, mu, atol=1e-1))
betaHat = np.sqrt(np.var(r, axis=0)/2.)
assert(np.allclose(betaHat, beta, atol=1e-1))
def llh(cls, parameters: Dict[str, Tensor], data: tf.Tensor) -> Tensor:
beta = parameters["beta"]
mu = tf.zeros_like(beta)
llh = LaplaceAlgorithms.llh(parameters={"mu": mu, "beta": beta},
data=data)
return(llh)
def pdf(cls, parameters: Dict[str, Tensor], data: Tensor) -> Tensor:
beta = parameters["beta"]
mu = tf.zeros_like(beta)
pdf = LaplaceAlgorithms.pdf(parameters={"mu": mu, "beta": beta},
data=data)
return(pdf)
def sample(cls, parameters: Dict[str, Tensor], nSamples: Tensor) -> Tensor:
beta = parameters["beta"]
mu = tf.zeros_like(beta)
samples = LaplaceAlgorithms.sample(parameters={"mu": mu, "beta": beta},
nSamples=nSamples)
return(samples)