Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Oct 8, 2020

• Eliminate usage of deprecated MCMC seed handling.

  Show description for cfeae22
  

  brianwa84

  authored and

  tensorflower-gardener

  committed
  cfeae22

  Copy full SHA for cfeae22

Commits on Oct 7, 2020

• Fix up the sample_chain example.

  Show description for 6cdc725
  

  cheshire

  authored and

  tensorflower-gardener

  committed
  6cdc725

  Copy full SHA for 6cdc725

• Remove duplicate entries from tfp.experimental.mcmc.__all__.

  Show description for 983f152
  

  SiegeLordEx

  authored and

  tensorflower-gardener

  committed
  983f152

  Copy full SHA for 983f152

• Update documentation of Mixture to suggest preferring MixtureSameFamily when possible.

  Show description for c0ad7fa
  

  axch

  authored and

  tensorflower-gardener

  committed
  c0ad7fa

  Copy full SHA for c0ad7fa

• Stop doing an in-place update in sample_stats.py. This has issues with the

  Show description for ea5351d
  

  langmore

  authored and

  tensorflower-gardener

  committed
  ea5351d

  Copy full SHA for ea5351d

• Fix composite tensor behavior for ScaleMatvecDiag and ScaleMatvecTriL. The issue was that the scale_diag and scale_tril parameters of each were not correctly interpreted as "tensors" for the compos…

  Show description for 283f014
  

  Googler

  authored and

  tensorflower-gardener

  committed
  283f014

  Copy full SHA for 283f014

• Drop usage of deprecated symbol.

  Show description for 7700018
  

  brianwa84

  authored and

  tensorflower-gardener

  committed
  7700018

  Copy full SHA for 7700018

• Internal change.

  Show description for efc0d4b
  

  brianwa84

  authored and

  tensorflower-gardener

  committed
  efc0d4b

  Copy full SHA for efc0d4b

• Ensure tfd.Sample broadcasts to full sample shape on computing log_prob.

  Show description for c36333c
  

  brianwa84

  authored and

  tensorflower-gardener

  committed
  c36333c

  Copy full SHA for c36333c

• BUGFIX: Must specify a default name for WishartLinearOperator, else

  Show description for 1c88e63
  

  langmore

  authored and

  tensorflower-gardener

  committed
  1c88e63

  Copy full SHA for 1c88e63

• Fix typo in Bates.log_prob input sanitization.

  Show description for 8684dec
  

  axch

  authored and

  tensorflower-gardener

  committed
  8684dec

  Copy full SHA for 8684dec

• Add GradientBasedTrajectoryLengthAdaptation.

  Show description for cadd4df
  

  SiegeLordEx

  authored and

  tensorflower-gardener

  committed
  cadd4df

  Copy full SHA for cadd4df

• Add tests for make_asvi_trainable_variables

  Show description for 5852dc3
  

  kateslin

  authored and

  tensorflower-gardener

  committed
  5852dc3

  Copy full SHA for 5852dc3

• Add Lorenz system variant where the global scale parameters are unknown.

  Show description for ba04db5
  

  davmre

  authored and

  tensorflower-gardener

  committed
  ba04db5

  Copy full SHA for ba04db5

• Fix tfb.Permute under jax.jit.

  Show description for 6dc8279
  

  SiegeLordEx

  authored and

  tensorflower-gardener

  committed
  6dc8279

  Copy full SHA for 6dc8279

Commits on Oct 6, 2020

• Make the experimental_default_event_space_bijector method public. The API remains experimental and subject to change.

  Show description for 758a89f
  

  brianwa84

  authored and

  tensorflower-gardener

  committed
  758a89f

  Copy full SHA for 758a89f

• Add the Parabolic (Epanechnikov) kernel.

  Show description for bd893cb
  

  brianwa84

  authored and

  tensorflower-gardener

  committed
  bd893cb

  Copy full SHA for bd893cb

• Add Brownian motion variant where the global scale parameters are unknown.

  Show description for d4c22de
  

  davmre

  authored and

  tensorflower-gardener

  committed
  d4c22de

  Copy full SHA for d4c22de

• Remove spurious prior sampling from STS parameter initialization.

  Show description for 81d6c60
  

  davmre

  authored and

  tensorflower-gardener

  committed
  81d6c60

  Copy full SHA for 81d6c60

• Merge pull request tensorflow#1049 from ksachdeva:rethinking-2nd-edition

  Show description for 86c051c
  tensorflower-gardener

  committed
  86c051c

  Copy full SHA for 86c051c

• Split assert_integer_form utility according to its two distinct purposes.

  Show description for f2bbae0
  

  axch

  authored and

  tensorflower-gardener

  committed
  f2bbae0

  Copy full SHA for f2bbae0

• Improve the speed of the Gamma sampler by replacing the nested loop rejection sampler with a single loop. Relaxed tolerance of testSampleGammaLowConcentration to alleviate sensitivity to random seeds.

  Show description for b040203
  

  ursk

  authored and

  tensorflower-gardener

  committed
  b040203

  Copy full SHA for b040203

• Use static shape utilities in LGSSM for JAX omni-staging compatibility.

  Show description for a68eaa8
  

  davmre

  authored and

  tensorflower-gardener

  committed
  a68eaa8

  Copy full SHA for a68eaa8

Commits on Oct 5, 2020

• Block tfp.experimental.substrates.numpy.experimental from being documented.

  Show description for 3d296c8
  

  yashk2810

  authored and

  tensorflower-gardener

  committed
  3d296c8

  Copy full SHA for 3d296c8

• Try to simplify the explanation of "inverse scale" in

  Show description for 06d8ac1
  

  langmore

  authored and

  tensorflower-gardener

  committed
  06d8ac1

  Copy full SHA for 06d8ac1

• Loosen test tolerances.

  Show description for 7d758c3
  

  ColCarroll

  authored and

  tensorflower-gardener

  committed
  7d758c3

  Copy full SHA for 7d758c3

Commits on Oct 3, 2020

• Do not generate docs for tfp.exprimental.substrates.jax.experimental

  Show description for f185c85
  

  yashk2810

  authored and

  tensorflower-gardener

  committed
  f185c85

  Copy full SHA for f185c85

Commits on Oct 2, 2020

• Fix error in preconditioned HMC sampler.

  Show description for 17af734
  

  ColCarroll

  authored and

  tensorflower-gardener

  committed
  17af734

  Copy full SHA for 17af734

• Add even_odd_swap_proposal_fn to __all__.

  Show description for 23f81f3
  

  langmore

  authored and

  tensorflower-gardener

  committed
  23f81f3

  Copy full SHA for 23f81f3

• untempered_log_prob_fn added as __init__ kwarg to ReplicaExchangeMC Kernel.

  Show description for 2f11f1f
  

  langmore

  authored and

  tensorflower-gardener

  committed
  2f11f1f

  Copy full SHA for 2f11f1f

• Replace the current generic log-concave discrete sampler used for binomial with a more specialized sampler ported from TF C++.

  Show description for e8dc108
  

  brianwa84

  authored and

  tensorflower-gardener

  committed
  e8dc108

  Copy full SHA for e8dc108

• Touchups from running NoNansInSampleTest with more computation.

  Show description for dbfc7fb
  

  axch

  authored and

  tensorflower-gardener

  committed
  dbfc7fb

  Copy full SHA for dbfc7fb

• Add a Hypothesis test that dist.sample does not emit nan values.

  Show description for 657398d
  

  axch

  authored and

  tensorflower-gardener

  committed
  657398d

  Copy full SHA for 657398d

• [Oryx] Fix use of jax.core.literalable_types

  Show description for 900c08d
  

  sharadmv

  authored and

  tensorflower-gardener

  committed
  900c08d

  Copy full SHA for 900c08d

• Inference Gym: Add Lorenz System target

  Show description for aaed4d3
  

  sharadmv

  authored and

  tensorflower-gardener

  committed
  aaed4d3

  Copy full SHA for aaed4d3

Pagination

Previous

Next