def sample_likelihood(self, zs, n): """x | z ~ p(x | z)""" out = [] for s in range(zs.shape[0]): out += [{'x': bernoulli.rvs(zs[s, :], size=n).reshape((n, ))}] return out
def sample_likelihood(self, zs, size): """x | z ~ p(x | z)""" out = np.zeros((zs.shape[0], size)) for s in range(zs.shape[0]): out[s,:] = bernoulli.rvs(zs[s,:], size=size).reshape((size,)) return out
def sample_likelihood(self, zs, size): """x | z ~ p(x | z)""" out = [] for s in range(zs.shape[0]): out += [{'x': bernoulli.rvs(zs[s, :], size=size).reshape((size,))}] return out
def sample_likelihood(self, zs, size): """x | z ~ p(x | z)""" out = np.zeros((zs.shape[0], size)) for s in range(zs.shape[0]): out[s, :] = bernoulli.rvs(zs[s, :], size=size) return out
def sample(self, size=1): """z ~ q(z | lambda)""" p = self.p.eval() z = np.zeros((size, self.num_vars)) for d in range(self.num_vars): z[:, d] = bernoulli.rvs(p[d], size=size) return z
def sample(self, size, sess): """z ~ q(z | lambda)""" p = sess.run(self.p) z = np.zeros(size) for d in range(self.num_vars): z[:, d] = bernoulli.rvs(p[d], size=size[0]) return z
def sample(self, size=1, sess=None): """z ~ q(z | lambda)""" p = sess.run(self.p) z = np.zeros((size, self.num_vars)) for d in range(self.num_vars): z[:, d] = bernoulli.rvs(p[d], size=size) return z
def sample(self, size=1): """x ~ p(x | params)""" p = self.p.eval() x = np.zeros((size, self.num_vars)) for d in range(self.num_vars): x[:, d] = bernoulli.rvs(p[d], size=size) return x
def sample(self, size=1): p = self.p.eval() return bernoulli.rvs(p, size=size)
def np_sample(p): # get `size` from lexical scoping return bernoulli.rvs(p, size=n).astype(np.float32)
def _test(p, size): val_est = bernoulli.rvs(p, size=size).shape val_true = (size, ) + np.asarray(p).shape assert val_est == val_true