Robuta

Computes log probability of actions given distribution.

Half-Student's t distribution.

Computes Dawson's integral element-wise.

Computes the scaled complementary error function exp(x**) * erfc(x).

An unconditional jump.

Dormand-Prince explicit solver for non-stiff ODEs.

Computes rhs^T matrix^-1 rhs, where matrix is HPSD.

Concatenates chol @ chol.T with additional rows and columns.

Exact KL Divergence.

Implementation of the modified Gram-Schmidt orthonormalization algorithm.

Pustaka untuk menggabungkan model probabilistik dan pembelajaran mendalam pada perangkat keras modern (TPU, GPU) untuk ilmuwan data, ahli statistik, peneliti...

Computes Y = g(X) = Abs(X), element-wise.

Augments doc on tfd.kl_divergence, EachDist.kl_divergence/cross_entropy.

Posterior predictive Normal distribution w. conjugate prior on the mean.

Computes Owen's T function of h and a element-wise.

Reduces the input tensor along the given axis using Kahan summation.

A mixture of a point-mass and another distribution.

Context object for auto-batching multiple Python functions together.

Helper which processes Tensors to tuples in standard form.

Linear 1-D interpolation on a regular (constant spacing) grid.

Runs one step of Metropolis-adjusted Langevin algorithm.

Densely-connected layer class with Flipout estimator.

TensorFlow Probability experimental python utilities.

Exponential distribution.

Construct predictive distribution over future observations.

Computes the inverse softplus, i.e., x = softplus_inverse(softplus(x)).

The Pearson Csiszar-function in log-space.

NegativeBinomial(total_count, probs=mean / (mean + total_count)).

Get the KL-divergence KL(distribution_a || distribution_b).

Compute g(X) = X ** power; where X is a non-negative real number.

The Batch-Reshaping distribution.

Get the KL-divergence KL(distribution_a || distribution_b).

Decompose a forecast distribution into contributions from each component.

Computes a conjugate normal posterior for a Bayesian linear regression.

The Cauchy distribution with location loc and scale scale.

Extracts a set of the variable names referenced by the node in question.

A function subject to auto-batching, callable with FunctionCallOp.

Compute Y = g(X) = (1 - exp(-rate * X)) * exp(-c * exp(-rate * X)).

A library to combine probabilistic models and deep learning on modern hardware (TPU, GPU) for data scientists, statisticians, ML researchers, and practitioners.

The finite discrete distribution.

Computes log|det(matrix)|, where matrix is a HPSD matrix.

Tools for probabilistic reasoning in TensorFlow.

Computes the power of one value to another.

Base class for stopping rules.

Methods and objectives for variational inference.

Condition to stop when all batch members have converged or failed.

Bijector that computes Y = tanh(X), therefore Y in (-1, 1).

TensorFlow Probability python utilities.

He normal initializer.

Map various PRNG seed flavors to a seed Tensor.

Bijector constructed from custom callables.

Automagically generate CompositeTensor behavior for cls.

Abstract base class for positive semi-definite kernel functions.

State space model for an autoregressive moving average process.

Tools for probabilistic reasoning in TensorFlow.

Returns a tensor with a length 1 axis inserted at index axis.

Estimate a lower bound on effective sample size for each independent chain.

The Laplace distribution with location loc and scale parameters.