def test_single_categorical_col_samples(model):
n = 1000
x = [int(y) for y in mu.sample([model], [0], n=n).flatten()]
cts = np.bincount(x, minlength=3)
assert cts[2] > cts[1]
assert cts[0] > cts[2]
def test_single_categorical_col_samples(model):
n = 1000
x = [int(y) for y in mu.sample([model], [0], n=n).flatten()]
cts = np.bincount(x, minlength=3)
assert cts[2] > cts[1]
assert cts[0] > cts[2]
def sample(self, cols, given=None, n=1):
""" Draw samples from cols
Parameters
----------
cols : list(index)
List of columns from which to jointly draw.
given : list(tuple(int, value))
List of column-value tuples that specify conditions
n : int
The number of samples to draw
Examples
--------
Draw whether an animal is fast and agile given that it is bulbous.
>>> engine = Engine.load('examples/zoo.bcmodels')
>>> engine.sample(['fast', 'agile'], given=[('bulbous': 1]))
"""
# FIXME: make sure that given does not contain columns from cols
if given is not None:
given = du.convert_given(given, self._dtypes, self._converters)
col_idxs = [self._converters['col2idx'][col] for col in cols]
data_out = mu.sample(self._models, col_idxs, given=given, n=n)
x = du.convert_data(data_out,
cols,
self._dtypes,
self._converters,
to_val=True)
if x.shape == (
1,
1,
):
return x[0, 0]
elif x.shape[0] == 1:
return x[0, :]
else:
return x
def entropy(self, col, n_samples=500):
""" The entropy of a column.
Notes
-----
Returns differential entropy for continuous feature (obviously).
Parameters
----------
col : indexer
The name of the column
n_samples : int
The number of samples to use for the Monte Carlo approximation
(if `col` is categorical).
Returns
-------
h : float
The entropy of `col`.
"""
col_idx = self._converters['col2idx'][col]
dtype = self._dtypes[col_idx]
# Unless x is enumerable (is categorical), we approximate h(x) using
# an importance sampling extimate of h(x) using p(x) as the importance
# function.
if dtype == 'categorical':
k = self._distargs[col_idx][0]
x = np.array([[i] for i in range(k)])
logps = mu.probability(x, self._models, (col_idx, ))
assert logps.shape == (k, )
h = -np.sum(np.exp(logps) * logps)
else:
x = mu.sample(self._models, (col_idx, ), n=n_samples)
logps = mu.probability(x, self._models, (col_idx, ))
h = -np.sum(logps) / n_samples
return h
def entropy(self, col, n_samples=500):
""" The entropy of a column.
Notes
-----
Returns differential entropy for continuous feature (obviously).
Parameters
----------
col : indexer
The name of the column
n_samples : int
The number of samples to use for the Monte Carlo approximation
(if `col` is categorical).
Returns
-------
h : float
The entropy of `col`.
"""
col_idx = self._converters['col2idx'][col]
dtype = self._dtypes[col_idx]
# Unless x is enumerable (is categorical), we approximate h(x) using
# an importance sampling extimate of h(x) using p(x) as the importance
# function.
if dtype == 'categorical':
k = self._distargs[col_idx][0]
x = np.array([[i] for i in range(k)])
logps = mu.probability(x, self._models, (col_idx,))
assert logps.shape == (k,)
h = -np.sum(np.exp(logps)*logps)
else:
x = mu.sample(self._models, (col_idx,), n=n_samples)
logps = mu.probability(x, self._models, (col_idx,))
h = -np.sum(logps) / n_samples
return h
def sample(self, cols, given=None, n=1):
""" Draw samples from cols
Parameters
----------
cols : list(index)
List of columns from which to jointly draw.
given : list(tuple(int, value))
List of column-value tuples that specify conditions
n : int
The number of samples to draw
Examples
--------
Draw whether an animal is fast and agile given that it is bulbous.
>>> engine = Engine.load('examples/zoo.bcmodels')
>>> engine.sample(['fast', 'agile'], given=[('bulbous': 1]))
"""
# FIXME: make sure that given does not contain columns from cols
if given is not None:
given = du.convert_given(given, self._dtypes, self._converters)
col_idxs = [self._converters['col2idx'][col] for col in cols]
data_out = mu.sample(self._models, col_idxs, given=given, n=n)
x = du.convert_data(data_out, cols, self._dtypes, self._converters,
to_val=True)
if x.shape == (1, 1,):
return x[0, 0]
elif x.shape[0] == 1:
return x[0, :]
else:
return x
