What's Changed

• LargeBatchVariationalStrategy caches triangular solves for test time by @jacobrgardner in https://github.com/cornellius-gp/gpytorch/pull/2698

Full Changelog: https://github.com/cornellius-gp/gpytorch/compare/v1.15...v1.15.1

What's Changed

• Speed Up Variational Strategy by @kayween in https://github.com/cornellius-gp/gpytorch/pull/2692
• Fix relative import in vlarge batch ariatinal strategy tests by @Balandat in https://github.com/cornellius-gp/gpytorch/pull/2693
• Poisson and negative binomial by @StefanoDamato in https://github.com/cornellius-gp/gpytorch/pull/2690

New Contributors

• @StefanoDamato made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2690

Full Changelog: https://github.com/cornellius-gp/gpytorch/compare/v1.14.3...v1.15

What's Changed

• Remove unused eps parameter from relevant kernels by @Thomas-Christie in https://github.com/cornellius-gp/gpytorch/pull/2675
• fix docstring in MultitaskMultivariateNormal by @kayween in https://github.com/cornellius-gp/gpytorch/pull/2680
• type hint fixes Iterable[Kernel] -> Kernel by @kayween in https://github.com/cornellius-gp/gpytorch/pull/2678
• RFF/linear prediction strategy fix (take 2) by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2681
• Fix for #2674 - Corrected sizes for alpha in RQKernel when using Deep GPs by @mrlj-hash in https://github.com/cornellius-gp/gpytorch/pull/2677

New Contributors

• @Thomas-Christie made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2675
• @kayween made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2680
• @mrlj-hash made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2677

Full Changelog: https://github.com/cornellius-gp/gpytorch/compare/v1.14.2...v1.14.3

Full Changelog: https://github.com/cornellius-gp/gpytorch/compare/v1.14.1...v1.14.2

This bugfix reverts a breaking change introduced by #2671.

What's Changed

• Ensure LinaerKernel stores ard_num_dims property. by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2635
• Export "constraints" submodule from package root by @chrisyeh96 in https://github.com/cornellius-gp/gpytorch/pull/2656
• [Docs] Fixes #2569: Incorrect Matern docs by @samjenner1 in https://github.com/cornellius-gp/gpytorch/pull/2657
• Loosen numeric tolerances in certain unit tests by @esantorella in https://github.com/cornellius-gp/gpytorch/pull/2661
• Replace deprecated np.Inf with np.inf by @saitcakmak in https://github.com/cornellius-gp/gpytorch/pull/2663
• Implement Hadamard Gaussian noise by @TobyBoyne in https://github.com/cornellius-gp/gpytorch/pull/2481
• Update precommit by @sdaulton in https://github.com/cornellius-gp/gpytorch/pull/2665
• Supporting passing full train inputs, rather than only supporting task indices to in HadamardGaussianLikelihood by @sdaulton in https://github.com/cornellius-gp/gpytorch/pull/2664
• RFF/linear prediction strategy fix by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2671

New Contributors

• @samjenner1 made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2657
• @TobyBoyne made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2481

Full Changelog: https://github.com/cornellius-gp/gpytorch/compare/v1.14...v1.14.1

What's Changed

• DOC: fix a typo, Hyperparamters -> Hyperparameters by @partev in https://github.com/cornellius-gp/gpytorch/pull/2588
• Allow for dimension-specific variance in linear kernel by @sdaulton in https://github.com/cornellius-gp/gpytorch/pull/2593
• Fix bug #2606: Setting mixing_weights=False in SoftmaxLikelihood still adds the learnable parameter W by @DiegoFreitasH in https://github.com/cornellius-gp/gpytorch/pull/2607
• Add typing annotations to gpytorch.Module by @chrisyeh96 in https://github.com/cornellius-gp/gpytorch/pull/2608
• Improve typing annotations for linear kernel by @chrisyeh96 in https://github.com/cornellius-gp/gpytorch/pull/2613
• Remove unused imports in GPyTorch Fully Bayesian example by @chrisyeh96 in https://github.com/cornellius-gp/gpytorch/pull/2612
• gpytorch.module: fix typing annotation by @chrisyeh96 in https://github.com/cornellius-gp/gpytorch/pull/2611
• Remove deprecated lazy tensor by @saitcakmak in https://github.com/cornellius-gp/gpytorch/pull/2615
• Support broadcasting batch shapes in MTMVN.from_independent_mvns by @saitcakmak in https://github.com/cornellius-gp/gpytorch/pull/2621
• Fix typo in Pyro_GPyTorch_High_Level.ipynb by @williamjsdavis in https://github.com/cornellius-gp/gpytorch/pull/2574
• Fix error in IndexKernel docstring by @parrangoiz in https://github.com/cornellius-gp/gpytorch/pull/2625
• Implement MVN.unsqueeze by @saitcakmak in https://github.com/cornellius-gp/gpytorch/pull/2624
• Update MVN.expand() to support non-lazy MVN & reuse scale_tril where possible by @saitcakmak in https://github.com/cornellius-gp/gpytorch/pull/2623
• Unpin jaxtyping by @saitcakmak in https://github.com/cornellius-gp/gpytorch/pull/2626
• Detach new_covar_cache to enable JIT tracing of models after fantasization by @SaiAakash in https://github.com/cornellius-gp/gpytorch/pull/2605
• DOC: fix typos in likelihood.py by @chrisyeh96 in https://github.com/cornellius-gp/gpytorch/pull/2486
• sign fix for gradient blocks by @m-julian in https://github.com/cornellius-gp/gpytorch/pull/2598
• Start adding typing annotations to ExactGP by @chrisyeh96 in https://github.com/cornellius-gp/gpytorch/pull/2436
• Hotfix: fix .diagonal() calls for keops kernel matrices. by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2590
• Version bump (Python 3.10) by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2627

New Contributors

• @DiegoFreitasH made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2607
• @williamjsdavis made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2574
• @parrangoiz made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2625
• @SaiAakash made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2605

Full Changelog: https://github.com/cornellius-gp/gpytorch/compare/v1.13...v1.14

What's Changed

• main and develop branches by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2542
• Include jaxtyping to allow for Tensor/LinearOperator typehints with sizes. by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2543
• fix: replace deprecated scipy.integrate.cumtrapz with cumulative_trapezoid by @natsukium in https://github.com/cornellius-gp/gpytorch/pull/2545
• use common notation for normal distribution N(\mu, \sigma^2) by @partev in https://github.com/cornellius-gp/gpytorch/pull/2547
• fix broken link in Simple_GP_Regression.ipynb by @partev in https://github.com/cornellius-gp/gpytorch/pull/2546
• Deprecate last_dim_is_batch (bump PyTorch version to >= 2.0) by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2549
• Added ability for priors of transformed distributions to have their p… by @hvarfner in https://github.com/cornellius-gp/gpytorch/pull/2551
• Avoid unnecessary memory allocation for covariance downdate in SGPR prediction strategy by @JonathanWenger in https://github.com/cornellius-gp/gpytorch/pull/2559
• Fix VNNGP with batches by @LuhuanWu in https://github.com/cornellius-gp/gpytorch/pull/2375
• fix a typo by @partev in https://github.com/cornellius-gp/gpytorch/pull/2570
• fix a typo by @partev in https://github.com/cornellius-gp/gpytorch/pull/2571
• DOC: improve the formatting in the documentation by @partev in https://github.com/cornellius-gp/gpytorch/pull/2578

New Contributors

• @natsukium made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2545
• @hvarfner made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2551

Full Changelog: https://github.com/cornellius-gp/gpytorch/compare/v1.12...v1.13

What's Changed

• Minor patch to Matern covariances by @j-wilson in https://github.com/cornellius-gp/gpytorch/pull/2378
• Fix error messages for ApproximateGP.get_fantasy_model by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2374
• Fix lazy kernel slicing when there are multiple outputs by @douglas-boubert in https://github.com/cornellius-gp/gpytorch/pull/2376
• Fix training status of noise model of HeteroskedasticNoise after exceptions by @fjzzq2002 in https://github.com/cornellius-gp/gpytorch/pull/2382
• Stop rbf_kernel_grad and rbf_kernel_gradgrad creating the full covariance matrix unnecessarily by @douglas-boubert in https://github.com/cornellius-gp/gpytorch/pull/2388
• Likelihood bugfix by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2395
• Update RTD configuration, and linear_operator requirement. by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2399
• Better support for missing labels by @Turakar in https://github.com/cornellius-gp/gpytorch/pull/2288
• Fix latex of gradients in docs by @jlperla in https://github.com/cornellius-gp/gpytorch/pull/2404
• Skip the warning in gpytorch.lazy.__getattr__ if name starts with _ by @saitcakmak in https://github.com/cornellius-gp/gpytorch/pull/2423
• Fix KeOps regressions from #2296. by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2413
• Update index.rst by @mkomod in https://github.com/cornellius-gp/gpytorch/pull/2449
• python should also be a runtime dependency by @jaimergp in https://github.com/cornellius-gp/gpytorch/pull/2457
• fix a typo: cannonical -> canonical by @partev in https://github.com/cornellius-gp/gpytorch/pull/2461
• Update distributions.rst by @chrisyeh96 in https://github.com/cornellius-gp/gpytorch/pull/2487
• Fix flaky SVGP classification test by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2495
• DOC: Fix typo in docstring. by @johanneskopton in https://github.com/cornellius-gp/gpytorch/pull/2493
• fix a typo by @partev in https://github.com/cornellius-gp/gpytorch/pull/2464
• DOC: fix formatting issue in RFFKernel documentation by @partev in https://github.com/cornellius-gp/gpytorch/pull/2463
• DOC: fix broken formatting in leave_one_out_pseudo_likelihood.py by @partev in https://github.com/cornellius-gp/gpytorch/pull/2462
• ConstantKernel by @SebastianAment in https://github.com/cornellius-gp/gpytorch/pull/2511
• DOC: fix broken URL in periodic_kernel.py by @partev in https://github.com/cornellius-gp/gpytorch/pull/2513
• Bug: Exploit Structure in get_fantasy_strategy by @naefjo in https://github.com/cornellius-gp/gpytorch/pull/2494
• Matern52 grad by @m-julian in https://github.com/cornellius-gp/gpytorch/pull/2512
• Added optional kwargs to ExactMarginalLogLikelihood call by @rafaol in https://github.com/cornellius-gp/gpytorch/pull/2522
• Corrected configuration of exclude statements in pre-commit configuration by @JonathanWenger in https://github.com/cornellius-gp/gpytorch/pull/2541

New Contributors

• @douglas-boubert made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2376
• @fjzzq2002 made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2382
• @jlperla made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2404
• @mkomod made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2449
• @jaimergp made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2457
• @partev made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2461
• @chrisyeh96 made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2487
• @johanneskopton made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2493
• @naefjo made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2494
• @rafaol made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2522

Full Changelog: https://github.com/cornellius-gp/gpytorch/compare/v1.11...v1.12

What's Changed

• Fix solve_triangular(Tensor, LinearOperator) not supported in VNNGP by @Turakar in https://github.com/cornellius-gp/gpytorch/pull/2323
• Metrics fixes and cleanup by @JonathanWenger in https://github.com/cornellius-gp/gpytorch/pull/2325
• Lock down doc requirements to prevent RTD failures. by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2339
• Fix typos in multivariate_normal.py by @manuelhaussmann in https://github.com/cornellius-gp/gpytorch/pull/2331
• add Hamming IMQ kernel by @samuelstanton in https://github.com/cornellius-gp/gpytorch/pull/2327
• Use torch.cdist for dist by @esantorella in https://github.com/cornellius-gp/gpytorch/pull/2336
• Enable fantasy models for multitask GPs Reborn by @yyexela in https://github.com/cornellius-gp/gpytorch/pull/2317
• Clean up deprecation warnings by @saitcakmak in https://github.com/cornellius-gp/gpytorch/pull/2348
• More informative string representation of MultitaskMultivariateNormal distributions. by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2333
• Mean and kernel functions for first and second derivatives by @ankushaggarwal in https://github.com/cornellius-gp/gpytorch/pull/2235
• Bugfix: double added log noise prior by @LuisAugenstein in https://github.com/cornellius-gp/gpytorch/pull/2355
• Remove Module.getattr by @saitcakmak in https://github.com/cornellius-gp/gpytorch/pull/2359
• Remove num_outputs from IndependentModelList by @saitcakmak in https://github.com/cornellius-gp/gpytorch/pull/2360
• keops periodic and keops kernels unit tests by @m-julian in https://github.com/cornellius-gp/gpytorch/pull/2296
• Deprecate checkpointing by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2361

New Contributors

• @Turakar made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2323
• @manuelhaussmann made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2331
• @esantorella made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2336
• @yyexela made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2317
• @ankushaggarwal made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2235
• @LuisAugenstein made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2355

Full Changelog: https://github.com/cornellius-gp/gpytorch/compare/v1.10...v1.11

What's Changed

• Re-add pyro + torch_master check by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2241
• Fix silently ignored arguments in IndependentModelList by @saitcakmak in https://github.com/cornellius-gp/gpytorch/pull/2249
• fix bug in nearest_neighbor_variational_strategy by @LuhuanWu in https://github.com/cornellius-gp/gpytorch/pull/2243
• Move infinite interval bounds check into Interval constructor by @Balandat in https://github.com/cornellius-gp/gpytorch/pull/2259
• Use ufmt for code formatting and import sorting by @Balandat in https://github.com/cornellius-gp/gpytorch/pull/2262
• Update nearest_neighbors.py by @yw5aj in https://github.com/cornellius-gp/gpytorch/pull/2267
• Use raw strings to avoid "DeprecationWarning: invalid escape sequence" by @saitcakmak in https://github.com/cornellius-gp/gpytorch/pull/2282
• Fix handling of re-used priors by @Balandat in https://github.com/cornellius-gp/gpytorch/pull/2269
• Fix BernoulliLikelihood documentation by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2285
• gpytorch.settings.variational_cholesky_jitter can be set dynamically. by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2255
• Likelihood docs update by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2292
• Improve development/contributing documentation by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2293
• Use raw strings to avoid "DeprecationWarning: invalid escape sequence" by @saitcakmak in https://github.com/cornellius-gp/gpytorch/pull/2295
• Update SGPR notebook by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2303
• Update linear operator dependency to 0.4.0 by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2321

New Contributors

• @yw5aj made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2267

Full Changelog: https://github.com/cornellius-gp/gpytorch/compare/v1.9.1...v1.10

What's Changed

• Fix LMCVariationalStrategy example in docs by @adamjstewart in https://github.com/cornellius-gp/gpytorch/pull/2112
• Accept closure argument in NGD optimizer step by @dannyfriar in https://github.com/cornellius-gp/gpytorch/pull/2118
• Fix bug with Multitask DeepGP predictive variances. by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2123
• Autogenerate parameter types in documentation from python typehints by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2125
• Retiring deprecated versions ofpsd_safe_cholesky, NotPSDError, and assert_allclose by @SebastianAment in https://github.com/cornellius-gp/gpytorch/pull/2130
• fix custom dtype_value_context setting by @sdaulton in https://github.com/cornellius-gp/gpytorch/pull/2132
• Include linear operator in installation instructions by @saitcakmak in https://github.com/cornellius-gp/gpytorch/pull/2131
• Fixes HalfCauchyPrior by @feynmanliang in https://github.com/cornellius-gp/gpytorch/pull/2137
• Fix return type of Kernel.covar_dist by @Balandat in https://github.com/cornellius-gp/gpytorch/pull/2138
• Change variable name for better understanding by @findoctorlin in https://github.com/cornellius-gp/gpytorch/pull/2135
• Expose jitter by @hughsalimbeni in https://github.com/cornellius-gp/gpytorch/pull/2136
• Add HalfNormal prior distribution for non-negative variables. by @ZitongZhou in https://github.com/cornellius-gp/gpytorch/pull/2147
• Fix multitask/added_loss_term bugs in SGPR regression by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2121
• fix bugs in test half Cauchy prior. by @ZitongZhou in https://github.com/cornellius-gp/gpytorch/pull/2156
• Generalize RandomModule by @feynmanliang in https://github.com/cornellius-gp/gpytorch/pull/2164
• MMVN.to_data_independent_dist returns correct variance for non-interleaved MMVN distributions. by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2172
• Update MSLL in metrics.py by @jongwonKim-1997 in https://github.com/cornellius-gp/gpytorch/pull/2177
• Update multitask example notebook by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2190
• Fix exception message for missing kernel lazy kernel attribute by @dannyfriar in https://github.com/cornellius-gp/gpytorch/pull/2195
• Improving _sq_dist when x1_eq_x2 by @SebastianAment in https://github.com/cornellius-gp/gpytorch/pull/2204
• Fix docs/requirements.txt by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2206
• As per issue '#2175 [Docs] GP Regression With Uncertain Inputs'. by @corwinjoy in https://github.com/cornellius-gp/gpytorch/pull/2200
• Avoid evaluating kernel when adding jitter by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2189
• Avoid evaluating kernel in expand_batch by @dannyfriar in https://github.com/cornellius-gp/gpytorch/pull/2185
• Deprecating postprocess by @SebastianAment in https://github.com/cornellius-gp/gpytorch/pull/2205
• Make PiecewisePolynomialKernel GPU compatible by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2217
• Let LazyEvaluatedKernelTensor recall the grad state at instantiation by @SebastianAment in https://github.com/cornellius-gp/gpytorch/pull/2229
• Doc Update for Posterior Model Distribution and Posterior Predictive Distribution by @varunagrawal in https://github.com/cornellius-gp/gpytorch/pull/2230
• Fix 08_Advanced_Usage links by @st-- in https://github.com/cornellius-gp/gpytorch/pull/2240
• Add device property to Kernels, add unit tests by @Balandat in https://github.com/cornellius-gp/gpytorch/pull/2234
• pass **kwargs to ApproximateGP.call in DeepGPLayer by @IdanAchituve in https://github.com/cornellius-gp/gpytorch/pull/2224

New Contributors

• @dannyfriar made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2118
• @SebastianAment made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2130
• @feynmanliang made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2137
• @findoctorlin made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2135
• @hughsalimbeni made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2136
• @ZitongZhou made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2147
• @jongwonKim-1997 made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2177
• @corwinjoy made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2200
• @varunagrawal made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2230
• @st-- made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2240
• @IdanAchituve made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2224

Full Changelog: https://github.com/cornellius-gp/gpytorch/compare/v1.9.0...v1.9.1

Starting with this release, the LazyTensor functionality of GPyTorch has been pulled out into its own separate Python package, called linear_operator. Most users won't notice the difference (at the moment), but power users will notice a few changes.

If you have your own custom LazyTensor code, don't worry: this release is backwards compatible! However, you'll see a lot of annoying deprecation warnings 😄

LazyTensor -> LinearOperator

• All gpytorch.lazy.*LazyTensor classes now live in the linear_operator repo, and are now called linear_operator.operator.*LinearOperator.
  • For example, gpytorch.lazy.DiagLazyTensor is now linear_operator.operators.DiagLinearOperator
  • The only major naming change: NonLazyTensor is now DenseLinearOperator
• gpytorch.lazify and gpytorch.delazify are now linear_operator.to_linear_operator and linear_operator.to_dense, respectively.
• The _quad_form_derivative method has been renamed to _bilinear_derivative (a more accurate name!)
• LinearOperator method names now reflect their corresponding PyTorch names. This includes:
  • add_diag -> add_diagonal
  • diag -> diagonal
  • inv_matmul -> solve
  • symeig -> eigh and eigvalsh
• LinearOperator now has the mT property

torch_function functionality

LinearOperators are now compatible with the torch api! For example, the following code works:

diag_linear_op = linear_operator.operators.DiagLinearOperator(torch.randn(10))
torch.matmul(diag_linear_op, torch.randn(10, 2))  # returns a torch.Tensor!

Other files that have moved:

• gpytorch.functions - all of the core functions used by LazyTensors now live in the LinearOperator repo. This includes: diagonalization, dsmm, inv_quad, inv_quad_logdet, matmul, pivoted_cholesky, root_decomposition, solve (formally inv_matmul), and sqrt_inv_matmul
• gpytorch.utils - a few have moved to the LinearOperator repo. This includes: broadcasting, cholesky, contour_intergral_quad, getitem, interpolation, lanczos, linear_cg, minres, permutation, stable_pinverse, qr, sparse, SothcasticLQ, and toeplitz.

Full Changelog: https://github.com/cornellius-gp/gpytorch/compare/v1.8.1...v1.9.0

Bug fixes

• MultitaskMultivariateNormal: fix tensor reshape issue by @adamjstewart in https://github.com/cornellius-gp/gpytorch/pull/2081
• Fix handling of prior terms in ExactMarginalLogLikelihood by @saitcakmak in https://github.com/cornellius-gp/gpytorch/pull/2039
• Fix bug in preconditioned KISS-GP / Hadamard Multitask GPs by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2090
• Add constant_constraint to ConstantMean by @gpleiss in https://github.com/cornellius-gp/gpytorch/pull/2082

New Contributors

• @mone27 made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2076

Full Changelog: https://github.com/cornellius-gp/gpytorch/compare/v1.8.0...v1.8.1

Major Features

• add variational nearest neighbor GP by @LuhuanWu in https://github.com/cornellius-gp/gpytorch/pull/2026

New Contributors

• @adamjstewart made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2061
• @m-julian made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2054
• @ngam made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2059
• @LuhuanWu made their first contribution in https://github.com/cornellius-gp/gpytorch/pull/2026

Full Changelog: https://github.com/cornellius-gp/gpytorch/compare/v1.7.0...v1.8.0

Important: This release requires Python 3.7 (up from 3.6) and PyTorch 1.10 (up from 1.9)

New Features

• gpytorch.metrics module offers easy-to-use metrics for GP performance.(#1870) This includes:
  • gpytorch.metrics.mean_absolute_error
  • gpytorch.metrics.mean_squared_error
  • gpytorch.metrics.mean_standardized_log_loss
  • gpytorch.metrics.negative_log_predictive_density
  • gpytorch.metrics.quantile_coverage_error
• Large scale inference (using matrix-multiplication techniques) now implements the variance reduction scheme described in Wenger et al., ICML 2022. (#1836)
  • This makes it possible to use LBFGS, or other line search based optimization techniques, with large scale (exact) GP hyperparameter optimization.
• Variational GP models support online updates (i.e. “fantasizing new models). (#1874)
  • This utilizes the method described in Maddox et al., NeurIPS 2021
• Improvements to gpytorch.priors
  • New HalfCauchyPrior (#1961)
  • LKJPrior now supports sampling (#1737)

Minor Features

• Add LeaveOneOutPseudoLikelihood for hyperparameter optimization (#1989)
• The PeriodicKernel now supports ARD lengthscales/periods (#1919)
• LazyTensors (A) can now be matrix multiplied with tensors (B) from the left hand side (i.e. B x A) (#1932)
• Maximum Cholesky retries can be controlled through a setting (#1861)
• Kernels, means, and likelihoods can be pickled (#1876)
• Minimum variance for FixedNoiseGaussianLikelihood can be set with a context manager (#2009)

Bug Fixes

• Fix backpropagation issues with KeOps kernels (#1904)
• Fix broadcasting issues with lazily evaluated kernels (#1971)
• Fix batching issues with PolynomialKernel (#1977)
• Fix issues with PeriodicKernel.diag() (#1919)
• Add more informative error message when train targets and the train prior distribution mismatch (#1905)
• Fix issues with priors on ConstantMean (#2042)

This release contains several bug fixes and performance improvements.

New Features

• Variational multitask models can output a single task per input (rather than all tasks per input) (#1769)

Small fixes

• LazyTensor#to method more closely matches the torch Tensor API (#1746)
• Add type hints and exceptions to kernels to improve usability (#1802)

Performance

• Improve the speed of fantasy models (#1752)
• Improve the speed of solves and log determinants with KroneckerProductLazyTensor (#1786)
• Prevent explicit kernel evaluation when expanding a LazyTensor kernel (#1813)

Fixes

• Fix indexing bugs with kernels (#1802, #1819, #1828)
• Fix cholesky bugs on CUDA (#1848)
• Remove lines of code that generate warnings in PyTorch 1.9 (#1835)

New features

• Add gpytorch.kernels.PiecewisePolynomialKernel (#1738)
• Include ability to turn off diagonal correction for SGPR models (#1717)
• Include ability to cast LazyTensor to half and float types (#1726)

Performance improvements

• Specialty MVN log_prob method for Gaussians with sum-of-Kronecker covariances (#1674)
• Ability to specify devices when concatenating rows of LazyTensors (#1712)
• Improvements to LazyTensor symeig method (#1725)

Bug fixes

• Fix to computing batch sizes of kernels (#1685)
• Fix SGPR prediction when fast_computations flags are turned off (#1709)
• Improve stability of stable_qr function (#1714)
• Fix bugs with pyro integration for full Bayesian inference (#1721)
• num_classes in gpytorch.likelihoods.DirichletLikelihood should be an integer (#1728)

This release adds 2 new model classes, as well as a number of bug fixes:

• GPLVM models for unsupervised learning
• Polya-Gamma GPs for GP classification In addition, this release contains numerous improvements to SGPR models (that have also been included in prior bug-fix releases).

New features

• Add example notebook that demos binary classification with Polya-Gamma augmentation (#1523)
• New model class: Bayesian GPLVM with Stochastic Variational Inference (#1605)
• Periodic kernel handles multi-dimensional inputs (#1593)
• Add missing data gaussian likelihoods (#1668)

Performance

• Speed up SGPR models (#1517, #1528, #1670)

Fixes

• Fix erroneous loss for ExactGP multitask models (#1647)
• Fix pyro sampling (#1594)
• Fix initialize bug for additive kernels (#1635)
• Fix matrix multiplication of rectangular ZeroLazyTensor (#1295)
• Dirichlet GPs use true train targets not labels (#1641)

Various bug fixes, including

• Use current PyTorch functionality (#1611, #1586)
• Bug fixes to Lanczos factorization (#1607)
• Fixes to SGPR model (#1607)
• Various fixes to LazyTensor math (#1576, #1584)
• SmoothedBoxPrior has a sample method (#1546)
• Fixes to additive-structure models (#1582)
• Doc fixes {#1603)
• Fix to index kernel and LCM kernels (#1608, #1592)
• Fixes to KeOps bypass (#1609)

Fixes

• Simplify interface for 3+ layer DSPP models (#1565)
• Fix marginal log likelihood calculation for exact Bayesian inference w/ Pyro (#1571)
• Remove CG warning for small matrices (#1562)
• Fix Pyro cluster-multitask example notebook (#1550)
• Fix gradients for KeOps tensors (#1543)
• Ensure that gradients are passed through lazily-evaluated kernels (#1518)
• Fix bugs for models with batched fantasy observations (#1529, #1499)
• Correct default latent_dim value for LMC variational models (#1512)

New features

• Create gpytorch.utils.grid.ScaleToBounds utility to replace gpytorch.utils.grid.scale_to_bounds method (#1566)
• Fix skip connections in Deep GP example (#1531)
• Add fantasy point support for structured kernel interpolation models (#1545)

Documentation

• Add default values to all gpytorch.settings (#1564)
• Improve Hadamard multitask notebook (#1537)

Performance

• Speed up SGPR models (#1517, #1528)

This release includes many major speed improvements, especially to Kronecker-factorized multi-output models.

Performance improvements

• Major speed improvements for Kronecker product multitask models (#1355, #1430, #1440, #1469, #1477)
• Unwhitened VI speed improvements (#1487)
• SGPR speed improvements (#1493)
• Large scale exact GP speed improvements (#1495)
• Random Fourier feature speed improvements (#1446, #1493)

New Features

• Dirichlet Classification likelihood (#1484) - based on Milios et al. (NeurIPS 2018)
• MultivariateNormal objects have a base_sample_shape attribute for low-rank/degenerate distributions (#1502)

New documentation

• Tutorial for designing your own kernels (#1421)

Debugging utilities

• Better naming conventions for AdditiveKernel and ProductKernel (#1488)
• gpytorch.settings.verbose_linalg context manager for seeing what linalg routines are run (#1489)
• Unit test improvements (#1430, #1437)

Bug Fixes

• inverse_transform is applied to the initial values of constraints (#1482)
• psd_safe_cholesky obeys cholesky_jitter settings (#1476)
• fix scaling issue with priors on variational models (#1485)

Breaking changes

• MultitaskGaussianLikelihoodKronecker (deprecated) is fully incorporated in MultitaskGaussianLikelihood (#1471)

Fixes

• Spectral mixture kernels work with SKI (#1392)
• Natural gradient descent is compatible with batch-mode GPs (#1416)
• Fix prior mean in whitened SVGP (#1427)
• RBFKernelGrad has no more in-place operations (#1389)
• Fixes to ConstantDiagLazyTensor (#1381, #1385)

Documentation

• Include example notebook for multitask Deep GPs (#1410)
• Documentation updates (#1408, #1434, #1385, #1393)

Performance

• KroneckerProductLazyTensors use root decompositions of children (#1394)
• SGPR now uses Woodbury formula and matrix determinant lemma (#1356)

Other

• Delta distributions have an arg_constraints attribute (#1422)
• Cholesky factorization now takes optional diagonal noise argument (#1377)

This release primarily focuses on performance improvements, and adds contour integral quadrature based variational models.

Major Features

Variational models with contour integral quadrature

• Add an MVM-based approach to whitened variatiational inference (#1372)
• This is based on the work in Fast Matrix Square Roots with Applications to Gaussian Processes and Bayesian Optimization

Minor Features

Performance improvements

• Kronecker product models compute a deterministic logdet (faster than the Lanczos-based logdet) (#1332)
• Improve efficiency of KroneckerProductLazyTensor symeig method (#1338)
• Improve SGPR efficiency (#1356)

Other improvements

• SpectralMixtureKernel accepts arbitrary batch shapes (#1350)
• Variational models pass around arbitrary **kwargs to the forward method (#1339)
• gpytorch.settings context managers keep track of their default value (#1347)
• Kernel objects can be pickle-d (#1336)

Bug Fixes

• Fix requires_grad checks in gpytorch.inv_matmul (#1322)
• Fix reshaping bug for batch independent multi-output GPs (#1368)
• ZeroMean accepts a batch_shape argument (#1371)
• Various doc fixes/improvements (#1327, #1343, #1315, #1373)

This release includes the following fixes:

• Fix caching issues with variational GPs (#1274, #1311)
• Ensure that constraint bounds are properly cast to floating point types (#1307)
• Fix bug with broadcasting multitask multivariate normal shapes (#1312)
• Bypass KeOps for small/rectangular kernels (#1319)
• Fix issues with eigenvectors=False in LazyTensor#symeig (#1283)
• Fix issues with fixed-noise LazyTensor preconditioner (#1299)
• Doc fixes (#1275, #1301)

Major Features

New variational and approximate models

This release features a number of new and added features for approximate GP models:

• Linear model of coregionalization for variational multitask GPs (#1180)
• Deep Sigma Point Process models (#1193)
• Mean-field decoupled (MFD) models from "Parametric Gaussian Process Regressors" (Jankowiak et al., 2020) (#1179)
• Implement natural gradient descent (#1258)
• Additional non-conjugate likelihoods (Beta, StudentT, Laplace) (#1211)

New kernels

We have just added a number of new specialty kernels:

• gpytorch.kernels.GaussianSymmetrizedKLKernel for performing regression with uncertain inputs (#1186)
• gpytorch.kernels.RFFKernel (random Fourier features kernel) (#1172, #1233)
• gpytorch.kernels.SpectralDeltaKernel (a parametric kernel for patterns/extrapolation) (#1231)

More scalable sampling

• Large-scale sampling with contour integral quadrature from Pleiss et al., 2020 (#1194)

Minor features

• Ability to set amount of jitter added when performing Cholesky factorizations (#1136)
• Improve scalability of KroneckerProductLazyTensor (#1199, #1208)
• Improve speed of preconditioner (#1224)
• Add symeig and svd methods to LazyTensors (#1105)
• Add TriangularLazyTensor for Cholesky methods (#1102)

Bug fixes

• Fix initialization code for gpytorch.kernels.SpectralMixtureKernel (#1171)
• Fix bugs with LazyTensor addition (#1174)
• Fix issue with loading smoothed box priors (#1195)
• Throw warning when variances are not positive, check for valid correlation matrices (#1237, #1241, #1245)
• Fix sampling issues with Pyro integration (#1238)

Major features

• GPyTorch is compatible with PyTorch 1.5 (latest release)
• Several bugs with task-independent multitask models are fixed (#1110)
• Task-dependent multitask models are more batch-mode compatible (#1087, #1089, #1095)

Minor features

• gpytorch.priors.MultivariateNormalPrior has an expand method (#1018)
• Better broadcasting for batched inducing point models (#1047)
• LazyTensor repeating works with rectangular matrices (#1068)
• gpytorch.kernels.ScaleKernel inherits the active_dims property from its base kernel (#1072)
• Fully-bayesian models can be saved (#1076)

Bug Fixes

• gpytorch.kernels.PeriodicKernel is batch-mode compatible (#1012)
• Fix gpytorch.priors.MultivariateNormalPrior expand method (#1018)
• Fix indexing issues with LazyTensors (#1029)
• Fix constants with gpytorch.mlls.GammaRobustVariationalELBO (#1038, #1053)
• Prevent doubly-computing derivatives of kernel inputs (#1042)
• Fix initialization issues with gpytorch.kernels.SpectralMixtureKernel (#1052)
• Fix stability of gpytorch.variational.DeltaVariationalStrategy

Major New Features and Improvements

Each feature in this section comes with a new example notebook and documentation for how to use them -- check the new docs!

• Added support for deep gaussian processes (#564).
• KeOps integration has been added -- replace certain gpytorch.kernels.SomeKernel with gpytorch.kernels.keops.SomeKernel with KeOps installed, and run exact GPs on 100000+ data points (#812).
• Variational inference has undergone significant internal refactoring! All old variational objects should still function, but many are deprecated. (#903).
• Our integration with Pyro has been completely overhauled and is now much improved. For examples of interesting GP + Pyro models, see our new examples (#903).
• Our example notebooks have been completely reorganized, and our documentation surrounding them has been rewritten to hopefully provide a better tutorial to GPyTorch (#954).
• Added support for fully Bayesian GP modelling via NUTS (#918).

Minor New Features and Improvements

• GridKernel and GridInterpolationKernel now support rectangular grids (#888).
• Added cylindrical kernel (#577).
• Added polynomial kernel (#668).
• Added tutorials on basic usage (hyperparameters, saving/loading, etc) (#685).
• get_fantasy_model now supports batched models (#693).
• Added a prior_mode context manager that causes GP models to evaluate in prior mode (#707).
• Added linear mean (#676).
• Added horseshoe prior (#719).
• Added polynomial kernel with derivatives (#783).
• Fantasy model computations now use QR for solving least squares problems, improving numerical stability (#790).
• All legacy functions have been removed, in favor of new function format in PyTorch (#799).
• Added Newton Girard kernel (#821).
• GP predictions now automatically clear caches when backpropagating through them. Previously, if you wanted to train through a GP in eval mode, you had to clear the caches manually by toggling the GP back to train mode and then to eval mode again. This is no longer necessary (#916).
• Added rational quadratic kernel (#330)
• Switch to using torch.cholesky_solve and torch.logdet now that they support batch mode / backwards (#880)
• Better / less redundant parameterization for correlation matrices e.g. in IndexKernel (#912).
• Kernels now define __getitem__, which allows slicing batch dimensions (#782).
• Performance improvements in the small data regime, e.g. n < 2000 (#926).
• Increased the size of kernel matrix for which Cholesky is the default solve strategy to n=800 (#946).
• Added an option for manually specifying a different preconditioner for AddedDiagLazyTensor (#930).
• Added precommit hooks that enforce code style (#927).
• Lengthscales have been refactored, and kernels have an is_stationary attribute (#925).
• All of our example notebooks now get smoke tested by our CI.
• Added a deterministic_probes setting that causes our MLL computation to be fully deterministic when using CG+Lanczos, which improves L-BFGS convergence (#929).
• The use of the Woodbury formula for preconditioner computations is now fully replaced by QR, which improves numerical stability (#968).

Bug fixes

• Fix a type error when calling backward on gpytorch.functions.logdet (#711).
• Variational models now properly skip posterior variance calculations if the skip_posterior_variances context is active (#741).
• Fixed an issue with diag mode for PeriodicKernel (#761).
• Stability improvements for inv_softplus and inv_sigmoid (#776).
• Fix incorrect size handling in InterpolatedLazyTensor for rectangular matrices (#906)
• Fix indexing in IndexKernel for batch mode (#911).
• Fixed an issue where slicing batch mode lazy covariance matrices resulted in incorrect behavior (#782).
• Cholesky gives a better error when there are NaNs (#944).
• Use psd_safe_cholesky in prediction strategies rather than torch.cholesky (#956).
• An error is now raised if Cholesky is used with KeOps, which is not supported (#959).
• Fixed a bug where NaNs could occur during interpoilation (#971).
• Fix MLL computation for heteroskedastic noise models (#870).

A full list of bug fixes and features will be out with the 0.4 release.

This release addresses breaking changes in the recent PyTorch 1.2 release. Currently, GPyTorch will run on either PyTorch 1.1 or PyTorch 1.2.

A full list of new features and bug fixes will be coming soon in a GPyTorch 0.4 release.

New Features

• Implement kernel checkpointing, allowing exact GPs on up to 1M data points with multiple GPUs (#499)
• GPyTorch now supports hard parameter constraints (e.g. bounds) via the register_constraint method on Module (#596)
• All GPyTorch objects now support multiple batch dimensions. In addition to training b GPs simultaneously, you can now train a b1 x b2 matrix of GPs simultaneously if you so choose (#492, #589, #627)
• RBFKernelGrad now supports ARD (#602)
• FixedNoiseGaussianLikelihood offers a better interface for dealing with known observation noise values. WhiteNoiseKernel is now hard deprecated (#593)
• InvMatmul, InvQuadLogDet and InvQuad are now twice differentiable (#603)
• Likelihood has been redesigned. See the new documentation for details if you are creating custom likelihoods (#591)
• Better support for more flexible Pyro models. You can now define likelihoods of the form p(y|f, z) where f is a GP and z are arbitrary latent variables learned by Pyro (#591).
• Parameters can now be recursively initialized with full names, e.g. model.initialize(**{"covar_module.base_kernel.lengthscale": 1., "covar_module.outputscale": 1.}) (#484)
• Added ModelList and LikelihoodList for training multiple GPs when batch mode can't be used -- see example notebooks (#471)

Performance and stability improvements

• CG termination is now more tolerance based, and will much more rarely terminate without returning good solves. Furthermore, a warning is raised if it ever does that includes suggested courses of action. (#569)
• In non-ARD mode, RBFKernel and MaternKernel use custom backward implementations for performance (#517)
• Up to a 3x performance improvement in the regime where the test set is very small (#615)
• The noise parameter in GaussianLikelihood now has a default lower bound, similar to sklearn (#596)
• psd_safe_cholesky now adds successively increasing amounts of jitter rather than only once (#610)
• Variational inference initialization now uses psd_safe_cholesky rather than torch.cholesky to initialize with the prior (#610)
• The pivoted Cholesky preconditioner now uses a QR decomposition for its solve rather than the Woodbury formula for speed and stability (#617)
• GPyTorch now uses Cholesky for solves with very small matrices rather than CG, resulting in reduced overhead for that setting (#586)
• Cholesky can additionally be turned on manually for help debugging (#586)
• Kernel distance computations now use torch.cdist when on PyTorch 1.1.0 in the non-batch setting (#642)
• CUDA unit tests now default to using the least used available GPU when run (#515)
• MultiDeviceKernel is now much faster (#491)

Bug Fixes

• Fixed an issue with variational covariances at test time (#638)
• Fixed an issue where the training covariance wasn't being detached for variance computations, occasionally resulting in backward errors (#566)
• Fixed an issue where active_dims in kernels was being applied twice (#576)
• Fixes and stability improvements for MultiDeviceKernel (#560)
• Fixed an issue where fast_pred_var was failing for single training inputs (#574)
• Fixed an issue when initializing parameter values with non-tensor values (#630)
• Fixed an issue with handling the preconditioner log determinant value for MLL computation (#634)
• Fixed an issue where prior_dist was being cached for VI, which was problematic for pyro models (#599)
• Fixed a number of issues with LinearKernel, including one where the variance could go negative (#584)
• Fixed a bug where training inputs couldn't be set with set_train_data if they are currently None (#565)
• Fixed a number of bugs in MultitaskMultivariateNormal (#545, #553)
• Fixed an indexing bug in batch_symeig (#547)
• Fixed an issue where MultitaskMultivariateNormal wasn't interleaving rows correctly (#540)

Other

• GPyTorch is now fully Python 3.6, and we've begun to include static type hints (#581)
• Parameters in GPyTorch no longer have default singleton batch dimensions. For example, the default shape of lengthscale is now torch.Size([1]) rather than torch.Size([1, 1]) (#605)
• setup.py now includes optional dependents, reads requirements from requirements.txt, does not require torch if pytorch-nightly is installed (#495)

0.2.1

You can install GPyTorch via Anaconda (#463)

Speed and stability

• Kernel distances use the JIT for fast computations (#464)
• LinearCG uses the JIT for fast computations (#464)
• Improve the stability of computing kernel distances (#455)

Features

Variational inference improvements

• Sped up variational models by batching all matrix solves in one call (#454)
• Can use the same set of inducing points for batch variational GPs (#445)
• Whitened variational inference for improved convergence (#493)
• Variational log likelihoods for BernoulliLikelihood are computed with quadrature (#473)

Multi-GPU Gaussian processes

• Can train and test GPs by dividing the kernel onto multiple GPUs (#450)

GPs with derivatives

• Can define RBFKernels for observations and their derivatives (#462)

LazyTensors

• LazyTensors can broadcast matrix multiplication (#459)
• Can use @ sign for matrix multiplication with LazyTensors

GP-list

• Convenience methods for training/testing multiple GPs in a list (#471)

Other

• Added a gpytorch.settings.fast_computations feature to (optionally) use Cholesky-based inference (#456)
• Distributions define event shapes (#469)
• Can recursively initialize parameters on GP modules (#484)

Bugs

• Can initialize noise in GaussianLikelihood (#479)
• Fixed bugs in SGPR kernel (#487)

v0.1.1

Features

• Batch GPs, which previously were a feature, are now well-documented and much more stable (see docs)
• Can add "fantasy observations" to models.
• Option for exact marginal log likelihood and sampling computations (this is slower, but potentially useful for debugging) (gpytorch.settings.fast_computations)

Bug fixes

• Easier usage of batch GPs
• Reduce bugs in additive regression models

Stability of hyperparameters

• Hyperparameters taht are constrained to be positive (e.g. variance, lengthscale, etc.) are now parameterized throught the softplus function (log(1 + e^x)) rather than through the log function
• This dramatically improves the numerical stability and optimization of hyperparameters
• Old models that were trained with log parameters will still work, but this is deprecated.
• Inference now handles certain numerical floating point round-off errors more gracefully.

Various stability improvements to variational inference

Other changes

• GridKernel can be used for data that lies on a perfect grid.
• New preconditioner for LazyTensors.
• Use batched cholesky functions for improved performance (requires updating PyTorch)

New features

• Implement diagonal correction for basic variational inference, improving predictive variance estimates. This is on by default.
• LazyTensor._quad_form_derivative now has a default implementation! While custom implementations are likely to still be faster in many cases, this means that it is no longer required to implement a custom _quad_form_derivative when implementing a new LazyTensor subclass.

Bug fixes

• Fix a number of critical bugs for the new variational inference.
• Do some hyperparameter tuning for the SV-DKL example notebook, and include fancier NN features like batch normalization.
• Made it more likely that operations internally preserve the ability to perform preconditioning for linear solves and log determinants. This may have a positive impact on model performance in some cases.

Variational inference has been refactored

• Easier to experiment with different variational approximations
• Massive performance improvement for SV-DKL

Experimental Pyro integration for variational inference

• See the example Pyro notebooks

Lots of tiny bug fixes

(Too many to name, but everything should be better 😬)

Beta release

GPyTorch is now available on pip! pip install gpytorch.

Important! This release requires the preview build of PyTorch (>= 1.0). You should either build from source or install pytorch-nightly. See the PyTorch docs for specific installation instructions.

If you were previously using GPyTorch, see the migration guide to help you move over.

What's new

• Batch mode: it is possible to train multiple GPs simultaneously
• Improved multitask models

Breaking changes

• gpytorch.random_variables have been replaced by gpytorch.distributions. These build upon PyTorch distributions.
  • gpytorch.random_variables.GaussianRandomVariable -> gpytorch.distributions.MultivariateNormal.
  • gpytorch.random_variables.MultitaskGaussianRandomVariable -> gpytorch.distributions.MultitaskMultivariateNormal.

Utilities

• gpytorch.utils.scale_to_bounds is now gpytorch.utils.grid.scale_to_bounds

Kernels

• GridInterpolationKernel, GridKernel, InducingPointKernel - the attribute base_kernel_module has become base_kernel (for consistency)
• AdditiveGridInterpolationKernel no longer exists. Now use `AdditiveStructureKernel(GridInterpolationKernel(...))
• MultiplicativeGridInterpolationKernel no longer exists. Now use ProductStructureKernel(GridInterpolationKernel(...))`.

Attributes (n_* -> num_*)

• IndexKernel: n_tasks -> num_tasks
• LCMKernel: n_tasks -> num_tasks
• MultitaskKernel: n_tasks -> num_tasks
• MultitaskGaussianLikelihood: n_tasks -> num_tasks
• SoftmaxLikelihood: n_features -> num_features
• MultitaskMean: n_tasks -> num_tasks
• VariationalMarginalLogLikelihood: n_data -> num_data
• SpectralMixtureKernel: n_dimensions -> ard_num_dims, n_mixtures -> num_mixtures