Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Nov 23, 2019

• Fix Jax tfp.math gradient and generic tests

  Show description for a26e967
  

  sharadmv

  authored and

  tensorflower-gardener

  committed
  a26e967

  Copy full SHA for a26e967

• Fix Numpy math generic_test and gradient_test.

  Show description for 397ac0f
  

  sharadmv

  authored and

  tensorflower-gardener

  committed
  397ac0f

  Copy full SHA for 397ac0f

• Make tfd.TransformedDistribution tape-safe.

  Show description for 0612375
  

  rryan

  authored and

  tensorflower-gardener

  committed
  0612375

  Copy full SHA for 0612375

Commits on Nov 22, 2019

• Update doctring references to tf.random.{uniform,normal}.

  Show description for 2065e03
  

  axch

  authored and

  tensorflower-gardener

  committed
  2065e03

  Copy full SHA for 2065e03

• Make tfd.TransformedDistribution tape-safe.

  Show description for 658fe02
  

  emilyfertig

  authored and

  tensorflower-gardener

  committed
  658fe02

  Copy full SHA for 658fe02

• Fix memory leak in bijector cache (for py3).

  Show description for 2c8f338
  

  csuter

  authored and

  tensorflower-gardener

  committed
  2c8f338

  Copy full SHA for 2c8f338

• Suppress more known TF2 failures from Travis tests.

  Show description for 59cb6ad
  

  axch

  authored and

  tensorflower-gardener

  committed
  59cb6ad

  Copy full SHA for 59cb6ad

• Deprecate tfb.masked_autoregressive_default_template.

  Show description for b99d964
  

  jburnim

  authored and

  tensorflower-gardener

  committed
  b99d964

  Copy full SHA for b99d964

• Use the known dtype structure of JDs to intelligently convert values to Tensor.

  Show description for 45e4e51
  

  davmre

  authored and

  tensorflower-gardener

  committed
  45e4e51

  Copy full SHA for 45e4e51

Commits on Nov 21, 2019

• BREAKING CHANGE: Remove deprecated behavior of logits, probs properties.

  Show description for da402e0
  

  jburnim

  authored and

  tensorflower-gardener

  committed
  da402e0

  Copy full SHA for da402e0

• More accurate survival function and CDF for the generalized Pareto distribution.

  Show description for 6c1a516
  

  axch

  authored and

  tensorflower-gardener

  committed
  6c1a516

  Copy full SHA for 6c1a516

Commits on Nov 20, 2019

• Explicitly export files needed by other packages

  Show description for 9399b67
  

  Googler

  authored and

  tensorflower-gardener

  committed
  9399b67

  Copy full SHA for 9399b67

• Fix the seed used in linear_gaussian_ssm_test.testIIDNormalSample.

  Show description for 84cdabc
  

  axch

  authored and

  tensorflower-gardener

  committed
  84cdabc

  Copy full SHA for 84cdabc

Commits on Nov 19, 2019

• Suppress three problems from the Guitar build, pending investigation.

  Show description for d406716
  

  axch

  authored and

  tensorflower-gardener

  committed
  d406716

  Copy full SHA for d406716

• Re-enable CustomConvertToCompositeTensorTest.test_str_repr

  Show description for fba6f07
  

  Googler

  authored and

  tensorflower-gardener

  committed
  fba6f07

  Copy full SHA for fba6f07

• Standardize bijectors that are CDF, Quantile pairs as *CDF.

  Show description for 968eab9
  

  srvasude

  authored and

  tensorflower-gardener

  committed
  968eab9

  Copy full SHA for 968eab9

• Temporarily suppress GaussianProcessRegressionModel in stochastic_process_properties_test.

  Show description for e8025e7
  

  axch

  authored and

  tensorflower-gardener

  committed
  e8025e7

  Copy full SHA for e8025e7

• Update docstrings in tfp.mcmc.random_walk_metropolis to TF2.0 API.

  Show description for 71882e3
  

  junpenglao

  authored and

  tensorflower-gardener

  committed
  71882e3

  Copy full SHA for 71882e3

• Prevent NaN acceptance ratios from poisoning step size adaptation.

  Show description for 1f8a3e8
  

  davmre

  authored and

  tensorflower-gardener

  committed
  1f8a3e8

  Copy full SHA for 1f8a3e8

• Add moving average stats: tfp.stats.assign_moving_mean_variance and tfp.stats.assign_log_moving_mean_exp.

  Show description for c181c54
  

  Joshua V. Dillon

  authored and

  tensorflower-gardener

  committed
  c181c54

  Copy full SHA for c181c54

• Remove deprecated tfb.AutoregressiveLayer.

  Show description for c4058cf
  

  jburnim

  authored and

  tensorflower-gardener

  committed
  c4058cf

  Copy full SHA for c4058cf

Commits on Nov 18, 2019

• Added FFJORD bijector class.

  Show description for f5fbc4a
  

  kochkov92

  authored and

  tensorflower-gardener

  committed
  f5fbc4a

  Copy full SHA for f5fbc4a

• Correctly override is_calibrated.

  Show description for da60244
  

  SiegeLordEx

  authored and

  tensorflower-gardener

  committed
  da60244

  Copy full SHA for da60244

Commits on Nov 15, 2019

• Implement helpers for suppressing tests from the Guitar build, since

  Show description for c37c71e
  

  axch

  authored and

  tensorflower-gardener

  committed
  c37c71e

  Copy full SHA for c37c71e

• Remove default value from TestUtil.maybe_static; set it in scale_test.py

  Show description for ea33999
  

  csuter

  authored and

  tensorflower-gardener

  committed
  ea33999

  Copy full SHA for ea33999

• Fix broadcasting bug in MVN*.{stddev,variance}

  Show description for 61877cb
  

  csuter

  authored and

  tensorflower-gardener

  committed
  61877cb

  Copy full SHA for 61877cb

• Disable OSS-irrelevant BUILD config line on export

  Show description for 6b554cd
  

  csuter

  authored and

  tensorflower-gardener

  committed
  6b554cd

  Copy full SHA for 6b554cd

• Explicitly export files needed by other packages

  Show description for cf9c904
  

  Googler

  authored and

  tensorflower-gardener

  committed
  cf9c904

  Copy full SHA for cf9c904

• Explicitly export files needed by other packages

  Show description for f143217
  

  Googler

  authored and

  tensorflower-gardener

  committed
  f143217

  Copy full SHA for f143217

• In KLDivergenceRegularizer, fix small bug which was ignoring value of use_exact_kl when it's True. This allows us to use exact kl_divergence instead of estimating from samples in test_points_fn.

  Show description for d82e845
  

  bloops

  authored and

  tensorflower-gardener

  committed
  d82e845

  Copy full SHA for d82e845

• Update GP regression notebook to TF2, and demo marginalization with HMC. (take 2)

  Show description for 2c53440
  davmre

  committed
  2c53440

  Copy full SHA for 2c53440

• Support OrderedDict dtype in JointDistributionNamed.

  Show description for 1fc4dd6
  davmre

  committed
  1fc4dd6

  Copy full SHA for 1fc4dd6

• Update GP regression notebook to TF2, and demo marginalization with HMC.

  Show description for 01a381e
  davmre

  committed
  01a381e

  Copy full SHA for 01a381e

• Fix example formatting in distribution_util.pick_vector

  Show description for 048cf93
  

  csuter

  authored and

  davmre

  committed
  048cf93

  Copy full SHA for 048cf93

• Update GP regression notebook to TF2, and demo marginalization with HMC.

  Show description for 46ac2e3
  davmre

  committed
  46ac2e3

  Copy full SHA for 46ac2e3

Pagination

Previous

Next