|View source on GitHub|
Create a random variable for OrderedLogistic.
tfp.edward2.OrderedLogistic( *args, **kwargs )
See OrderedLogistic for more details.
Original Docstring for Distribution
Initialize Ordered Logistic distributions.
cutpoints: A floating-point
[B1, ..., Bb, K]where
b >= 0indicates the number of batch dimensions. Each entry is then a
K-length vector of cutpoints. The vector of cutpoints should be non-decreasing, which is only checked if
loc: A floating-point
[B1, ..., Bb]where
b >= 0indicates the number of batch dimensions. The entries represent the mean(s) of the latent logistic distribution(s). Different batch shapes for
locare permitted, with the distribution
tf.shape(loc[..., tf.newaxis] - cutpoints)[:-1]assuming the subtraction is a valid broadcasting operation.
dtype: The type of the event samples (default: int32).
Truedistribution parameters are checked for validity despite possibly degrading runtime performance. When
Falseinvalid inputs may silently render incorrect outputs.
True, statistics (e.g. mode) use the value "
NaN" to indicate the result is undefined. When
False, an exception is raised if one or more of the statistic's batch members are undefined.
strname prefixed to Ops created by this class.