a78c46f | Vincent Adam | 07 March 2019, 17:10:26 UTC | Fixing bad docstrings in conditionals (#952) | 07 March 2019, 17:10:26 UTC |
cb34808 | st-- | 06 March 2019, 17:21:44 UTC | Notebook fixes (1/...): intro, tips_and_tricks, basics/ (#948) Proof-reading of notebooks: intro, tips_and_tricks, and basics/{regression, classification, GPLVM}. Fixes a few links and typos elsewhere. | 06 March 2019, 17:21:44 UTC |
062c697 | Artem Artemev | 03 March 2019, 16:28:34 UTC | GPflow glossary document (#923) | 03 March 2019, 16:28:34 UTC |
f376dc2 | Marcin Tomczak | 01 March 2019, 20:57:31 UTC | Fix `fully_correlated_conditional_repeat` shapes (#945) | 01 March 2019, 20:57:31 UTC |
7996cf7 | Ayman Boustati | 25 February 2019, 11:28:16 UTC | Merge pull request #946 from GPflow/ayman/multioutput-notebook Clean up of multioutput notebook and link to other notebooks | 25 February 2019, 11:28:16 UTC |
dc7db32 | st-- | 25 February 2019, 11:06:18 UTC | Merge branch 'develop' into ayman/multioutput-notebook | 25 February 2019, 11:06:18 UTC |
8517439 | Ayman Boustati | 25 February 2019, 10:49:39 UTC | gone over review and implmented changes | 25 February 2019, 10:49:39 UTC |
5ef8813 | Simon Fothergill | 20 February 2019, 17:36:23 UTC | Added GPflows diagram to intro (and altered intro paragraph to explain what comes next) (#947) | 20 February 2019, 17:36:23 UTC |
8b14095 | Ayman Boustati | 20 February 2019, 12:34:14 UTC | clean up of multioutput notebook and link to other notebooks | 20 February 2019, 12:34:14 UTC |
5f037b4 | Artem Artemev | 19 February 2019, 17:10:16 UTC | Fix intro.ipynb rendering in nbviewer | 19 February 2019, 17:10:16 UTC |
c2c8da2 | Artem Artemev | 19 February 2019, 17:02:17 UTC | Remove empty cell from intro.ipynb | 19 February 2019, 17:02:17 UTC |
d3a0a8e | Artem Artemev | 19 February 2019, 16:56:03 UTC | README cleaning | 19 February 2019, 16:56:03 UTC |
0dec75e | st-- | 19 February 2019, 16:37:32 UTC | Revamp of GPflow documentation and notebooks (#924) | 19 February 2019, 16:37:32 UTC |
649b039 | Vincent Dutordoir | 15 February 2019, 13:32:55 UTC | Minor performance improvement: pass Cholesky of K to gauss_kl (#933) Add option to pass K_cholesky to kl_gauss | 15 February 2019, 13:32:55 UTC |
191a1c2 | Vincent Dutordoir | 04 February 2019, 10:57:34 UTC | Broadcasting of conditionals (#895) | 04 February 2019, 10:57:34 UTC |
771de3e | st-- | 01 February 2019, 18:20:08 UTC | Minor changes to Tips & Tricks notebook (#921) * minor changes to tips_and_tricks notebook * add tips_and_tricks to docs * add doc/README.md | 01 February 2019, 18:20:08 UTC |
765224b | Artem Artemev | 30 January 2019, 01:11:11 UTC | Tips & Tricks notebooks changes (#916) | 30 January 2019, 01:11:11 UTC |
0bf75f7 | st-- | 29 January 2019, 11:27:39 UTC | Fix AutoBuild metaclass to not screw up method signatures (#907) The AutoBuild metaclass that is present in all Parameterized subclasses (e.g. models, kernels, likelihoods etc.) overwrote the `__init__` function and thereby replaced the signature with `*args, **kwargs`, which meant that the API documentation and command-line/IDE auto-completion was broken. This PR makes use of the "decorator" package (slightly hacked) to restore the original signatures (while allowing autobuild to still be passed as an additional keyword argument). | 29 January 2019, 11:27:39 UTC |
5166d0f | Artem Artemev | 29 January 2019, 08:16:48 UTC | Merge branch 'develop' of github.com:GPflow/GPflow into develop | 29 January 2019, 08:16:48 UTC |
c6805dc | Artem Artemev | 29 January 2019, 08:16:40 UTC | Change github template | 29 January 2019, 08:16:40 UTC |
35e58c4 | Artem Artemev | 28 January 2019, 22:11:34 UTC | Update README.md | 28 January 2019, 22:11:34 UTC |
c613302 | Artem Artemev | 28 January 2019, 21:31:13 UTC | Add references to documentation and tips & tricks in README. | 28 January 2019, 21:31:13 UTC |
f8fc156 | Artem Artemev | 28 January 2019, 18:13:37 UTC | Results of hackathon 28 January (#914) | 28 January 2019, 18:13:37 UTC |
1c74ce4 | James Hensman | 28 January 2019, 17:13:41 UTC | Hackathon doc cleanup (#906) * Release v1.3.0 (#869) * cleaning up some of the .rst files into a flatter hierarchy * editing upper bound notebook so that headers don't interfere with the doc structure * some eidts to the GPR docstring and call signature * imrpving kernel docstrings * improving SGPR call signature * simplifying use of abc.abstractmethod * reinstating a very important period * Update gpflow/kernels.py Co-Authored-By: jameshensman <james.hensman@gmail.com> * improvements for kernel docstrings * typo fix in gpflow/models/gpr.py Co-Authored-By: jameshensman <james.hensman@gmail.com> * typo fix in gpflow/models/sgpr.py Co-Authored-By: jameshensman <james.hensman@gmail.com> * more kernel docstring improvements * Added module level descriptions of likelihoods and kernels, fixed table in intro, and added auto-detection of modules with auto-members * Fixed minor typos in the docs * improving TOC by setting trtee depth to 2 * some copyright edits * Moved documentation from .rst to module level documentation * including missing argument in SGPR init * Added mean function module level documentation, changed to unicode from LaTeX for readability in code | 28 January 2019, 17:13:41 UTC |
7e55422 | st-- | 28 January 2019, 16:32:15 UTC | Typo "NOSET" -> "NOTSET" in gpflowrc (#911) The available Python logging levels are as written now in the file: https://docs.python.org/3/library/logging.html#levels | 28 January 2019, 16:32:15 UTC |
f78ade1 | st-- | 28 January 2019, 15:17:56 UTC | Privatize core methods (#908) * remove deprecated full_name (replaced by pathname) * rename a bunch of core methods to be private to clean up API docs | 28 January 2019, 15:17:56 UTC |
ba8b7a5 | Alan Saul | 28 January 2019, 14:56:28 UTC | Update README to include local compilation of docs (#904) * Update README to include instructions for local compilation of docs | 28 January 2019, 14:56:28 UTC |
2696407 | st-- | 28 January 2019, 14:49:34 UTC | fix models maths documentation (#905) (#909) Some docstrings had \\ for LaTeX maths commands but had a r at the start, so they were escaped too much. One docstring was missing the r but had some single-\. Fixes those. | 28 January 2019, 14:49:34 UTC |
cc5693c | Vincent Dutordoir | 09 January 2019, 17:25:24 UTC | Hotfix fully_correlated_conditional (#899) | 09 January 2019, 17:25:24 UTC |
57f4c14 | Artem Artemev | 06 January 2019, 00:57:33 UTC | Change github issue templates. | 06 January 2019, 00:57:33 UTC |
c43d3f5 | Artem Artemev | 06 January 2019, 00:46:41 UTC | Issue templates | 06 January 2019, 00:46:41 UTC |
d8fc9c1 | Artem Artemev | 23 December 2018, 00:02:46 UTC | CircleCI config. Add version to workflow | 23 December 2018, 00:02:46 UTC |
880c35d | Artem Artemev | 22 December 2018, 23:49:51 UTC | CircleCI status badge | 22 December 2018, 23:49:51 UTC |
27ff38c | Artem Artemev | 22 December 2018, 16:38:02 UTC | Add names to run commands for circleci | 22 December 2018, 16:38:02 UTC |
d5be4ee | Artem Artemev | 21 December 2018, 13:15:28 UTC | CircleCI instead of TravisCI (#892) | 21 December 2018, 13:15:28 UTC |
7f0cac1 | Artem Artemev | 03 December 2018, 00:36:05 UTC | Configure travis to work with jobs (#888) | 03 December 2018, 00:36:05 UTC |
856193d | Artem Artemev | 02 December 2018, 08:53:44 UTC | Travis configuration. Remove pip -y option | 02 December 2018, 08:53:44 UTC |
132a194 | Artem Artemev | 02 December 2018, 08:43:29 UTC | Create conda env with yes option | 02 December 2018, 08:43:29 UTC |
44090f2 | Artem Artemev | 02 December 2018, 08:28:13 UTC | PIP install with auto yes | 02 December 2018, 08:28:13 UTC |
0bc4b16 | Artem Artemev | 02 December 2018, 00:50:29 UTC | Travis configuration (#887) | 02 December 2018, 00:50:29 UTC |
fdb8efb | Vincent Dutordoir | 28 November 2018, 19:37:13 UTC | Support TensorFlow 1.12 (#877) | 28 November 2018, 19:37:13 UTC |
18fa159 | Artem Artemev | 24 November 2018, 22:56:31 UTC | Debug Travis | 24 November 2018, 22:56:31 UTC |
0ab60c3 | Artem Artemev | 24 November 2018, 22:42:05 UTC | Debug Travis | 24 November 2018, 22:42:05 UTC |
66f5287 | Artem Artemev | 24 November 2018, 12:12:08 UTC | Remove printing in travis.sh | 24 November 2018, 12:12:08 UTC |
62132f8 | Artem Artemev | 24 November 2018, 12:05:14 UTC | Print travis information | 24 November 2018, 12:05:14 UTC |
6b4a3ab | Artem Artemev | 24 November 2018, 00:21:56 UTC | Add travis script | 24 November 2018, 00:21:56 UTC |
84abb1d | Artem Artemev | 23 November 2018, 16:43:51 UTC | Merge branch 'develop' of github.com:GPflow/GPflow into develop | 23 November 2018, 16:43:51 UTC |
5616e65 | Artem Artemev | 23 November 2018, 14:11:08 UTC | Travis script (#884) * Travis script for running notebook tests only in develop and master branches. | 23 November 2018, 14:11:08 UTC |
cada28a | Artem Artemev | 22 November 2018, 17:43:22 UTC | Merge branch 'develop' of github.com:GPflow/GPflow into develop | 22 November 2018, 17:43:22 UTC |
14423f0 | Mark van der Wilk | 14 November 2018, 20:53:23 UTC | Compute SGPR qu's autoflow method (#879) | 14 November 2018, 20:53:23 UTC |
9562b9b | Artem Artemev | 14 November 2018, 14:20:40 UTC | Merge branch 'develop' of github.com:GPflow/GPflow into develop | 14 November 2018, 14:20:40 UTC |
abd8587 | Mark van der Wilk | 06 November 2018, 14:03:46 UTC | Remove `.fixed` property from Parameter class. (#873) | 06 November 2018, 14:03:46 UTC |
703ccb1 | Artem Artemev | 19 October 2018, 16:42:14 UTC | Merge branch 'develop' of github.com:GPflow/GPflow into develop | 19 October 2018, 16:42:14 UTC |
fe82d3f | Artem Artemev | 19 October 2018, 16:38:49 UTC | Release version 1.3.0 (#870) | 19 October 2018, 16:38:49 UTC |
377ff5f | Artem Artemev | 19 October 2018, 16:05:12 UTC | Merge branch 'develop' of github.com:GPflow/GPflow into develop | 19 October 2018, 16:05:12 UTC |
9587e10 | Artem Artemev | 19 October 2018, 15:32:02 UTC | Change release version to 1.3.0 | 19 October 2018, 15:32:02 UTC |
63975b1 | Artem Artemev | 19 October 2018, 12:44:17 UTC | Release notes 1.3.0 (#868) | 19 October 2018, 12:44:17 UTC |
120b128 | Artem Artemev | 19 October 2018, 11:05:23 UTC | Update RELEASE.md | 19 October 2018, 11:05:23 UTC |
1707d7b | Artem Artemev | 06 October 2018, 18:12:52 UTC | Merge branch 'develop' of github.com:GPflow/GPflow into develop | 06 October 2018, 18:12:52 UTC |
3935d6b | Hugh Salimbeni | 28 September 2018, 16:00:19 UTC | Merge pull request #854 from GPflow/hughsalimbeni-patch-1 added link to orthogonally decoupled repo | 28 September 2018, 16:00:19 UTC |
185cc8f | st-- | 27 September 2018, 10:14:51 UTC | Merge branch 'develop' into hughsalimbeni-patch-1 | 27 September 2018, 10:14:51 UTC |
e8ba02e | ahsimb | 27 September 2018, 10:14:20 UTC | Fixing the problem with monitoring ScipyOptimizer (#856) Now lets the monitor context keep track of the optimiser to update optimiser's variables - necessary for obtaining correct traces (of hyperparameters, log-marginal likelihoods etc.) when using ScipyOptimizer. | 27 September 2018, 10:14:20 UTC |
f8deb4c | st-- | 27 September 2018, 09:24:45 UTC | clean up GPR docstrings (#859) | 27 September 2018, 09:24:45 UTC |
1196600 | Artem Artemev | 20 September 2018, 10:34:26 UTC | Merge branch 'develop' of github.com:GPflow/GPflow into develop | 20 September 2018, 10:34:26 UTC |
bb3d1f4 | st-- | 19 September 2018, 13:48:08 UTC | Clean up notebooks and documentation (#849) Minor code changes: - adds rich representation to Parameter(ized) objects so `m.as_pandas_table()` can be replaced by `m` for pretty-printing in jupyter notebooks - updates priors to have prettier printing (and more value checks) Notebooks - coreg_demo.ipynb: clean-up, typo fixes - models.ipynb: clean-up, now actually consistent with current (gpflow-1.0) architecture - kernels.ipynb: clean-up Documentation - updates various links in README.md (JH's webpage, readthedocs, tensorflow installation) - adds all currently existing notebooks to the documentation (under new section "advanced tutorials") Stand-alone .py demo scripts in doc/source/notebooks/ - makes them runnable with current gpflow version - adds "unmaintained" warning header | 19 September 2018, 13:48:08 UTC |
2a16f04 | Hugh Salimbeni | 16 September 2018, 16:30:26 UTC | added link to orthogonally decoupled repo pending arxiv/nips 2018 publication | 16 September 2018, 16:30:26 UTC |
83631b7 | st-- | 14 September 2018, 16:01:26 UTC | Update docs_require.txt to supported tensorflow version (#853) | 14 September 2018, 16:01:26 UTC |
b94500b | Artem Artemev | 14 September 2018, 15:35:12 UTC | Merge branch 'develop' of github.com:GPflow/GPflow into develop | 14 September 2018, 15:35:12 UTC |
fa20322 | Artem Artemev | 14 September 2018, 15:34:55 UTC | Docs use supported tensorflow version | 14 September 2018, 15:34:55 UTC |
460afc7 | st-- | 12 September 2018, 19:01:15 UTC | Sample conditional returns mean and var as well as samples, and generates more than one sample (#836) | 12 September 2018, 19:01:15 UTC |
dc0b987 | Artem Artemev | 09 September 2018, 22:18:18 UTC | Merge branch 'develop' of github.com:GPflow/GPflow into develop | 09 September 2018, 22:18:18 UTC |
c8b86d3 | Artem Artemev | 09 September 2018, 22:18:09 UTC | New line in docstrings of Stationary kernels | 09 September 2018, 22:18:09 UTC |
2475891 | Joshua George Albert | 05 September 2018, 15:12:03 UTC | fix active_dims=slice (#840) Fixes #839 | 05 September 2018, 15:12:03 UTC |
12c7656 | st-- | 31 August 2018, 17:57:46 UTC | add link to github.com/Joshuaalbert/heterogp in lieu of #735 (#773) * add link to github.com/Joshuaalbert/heterogp in lieu of #735 * remove link to gpflow-monitor now that it's in GPflow itself, ok'd by @markvdw | 31 August 2018, 17:57:46 UTC |
a2afcee | sanket-kamthe | 30 August 2018, 09:23:09 UTC | RBF Cross-Covariance (#754) Adds RBF cross covariance in expectations. Previous implementation defaulted to quadrature expectations for cross co-variance when kernel and feature were not the same. | 30 August 2018, 09:23:09 UTC |
0f6a5d3 | Joshua George Albert | 29 August 2018, 22:11:37 UTC | Added MixedKernelSeparateMof (#830) Adds a new feature MixedKernelSeparateMof (SharedIndependentMof subclass) for the most efficient code together with SeparateMixedMok. | 29 August 2018, 22:11:37 UTC |
fd6beaa | Hugh Salimbeni | 22 August 2018, 14:13:39 UTC | Support broadcasting (#829) Modifies the kernels and conditionals to support broadcasting over (arbitrarily many) extra leading dimensions, similar to most of the tensorflow ops. So e.g. kern.K(X) now supports X being shape `[S, ..., T, N, D]`, and returns `[S, ..., T, N, N]` | 22 August 2018, 14:13:39 UTC |
af89b6d | st-- | 22 August 2018, 10:26:46 UTC | Cleaning up stationary kernel implementations (#827) Stationary kernels are now implemented by K_r2 as a function of the scaled square distance or K_r as a function of the scaled Euclidean distance, see the docstrings for details. scaled_square_dist and scaled_euclid_dist are deprecated, and square_dist and euclid_dist are removed. RBF is now an alias for SquaredExponential. | 22 August 2018, 10:26:46 UTC |
51e17ae | ST John | 17 August 2018, 21:00:46 UTC | Merge branch 'master' of http://github.com/GPflow/GPflow into develop | 17 August 2018, 21:00:46 UTC |
2c4cf39 | Alexander G. de G. Matthews | 17 August 2018, 19:05:27 UTC | Adding widedeepnetworks repository link (#825) * Adding repository link * Correction to table | 17 August 2018, 19:05:27 UTC |
66aa1f9 | st-- | 16 August 2018, 10:30:04 UTC | Extended ndiagquad docstrings (#823) | 16 August 2018, 10:30:04 UTC |
e41d8f3 | Vincent Dutordoir | 13 August 2018, 23:08:14 UTC | Bugfix control dependencie in likelihoods (#821) Fixes a bug with control dependencies in likelihoods. | 13 August 2018, 23:08:14 UTC |
d9624ae | st-- | 06 August 2018, 09:49:37 UTC | remove references to six module (#816) GPflow no longer supports python 2 and hence references to the six module are redundant. | 06 August 2018, 09:49:37 UTC |
54a276a | st-- | 19 July 2018, 13:19:51 UTC | Parameter.trainable was not properly a property, and param.trainable = False did not work. Fixed by this commit. (#814) | 19 July 2018, 13:19:51 UTC |
f746253 | Vincent Dutordoir | 17 July 2018, 12:58:31 UTC | Fix bug in ndiag_mc re multi-dimensional kwargs (#813) | 17 July 2018, 12:58:31 UTC |
1db48f3 | Artem Artemev | 11 July 2018, 12:50:44 UTC | Update version.py file to 1.2.0 (#812) | 11 July 2018, 12:50:44 UTC |
707b195 | Artem Artemev | 11 July 2018, 11:21:33 UTC | Release notes 1.2.0 (#810) | 11 July 2018, 11:21:33 UTC |
1ff0a87 | Artem Artemev | 11 July 2018, 10:58:39 UTC | Optimizer's step_callback accepts step number (#811) | 11 July 2018, 10:58:39 UTC |
b78c1a8 | ahsimb | 10 July 2018, 11:13:25 UTC | Fixed discovered problem with the monitor, in particular with TensorBoard tasks (#807) | 10 July 2018, 11:13:25 UTC |
495b6eb | Mark van der Wilk | 28 June 2018, 12:59:50 UTC | Softmax fix (#804) * PEP8. * PEP8. * Added `SoftMax.conditional_variance()`. * SoftMax test now tests against `Bernoulli` likelihood, including `predict_mean_and_var`. * super * consistency * test comment * bugfix * fix tests | 28 June 2018, 12:59:50 UTC |
45fed42 | ahsimb | 28 June 2018, 11:57:28 UTC | Big re-work of monitor (#792) * New monitor initial commit * Fixed the monitor notebook (checking that the saving directory exists) * Addressed PR comments. * refactored environment * Attempt to avoid using X server when testing the monitor. * Added few more tests. | 28 June 2018, 11:57:28 UTC |
1851e0d | st-- | 27 June 2018, 11:19:26 UTC | Monte-Carlo likelihoods (#799) * add @markvdw's stochastic likelihood, including the softmax * initial MC likelihood * remove MonteCarloLikelihood base class from tests * fix test * var of predict_mean_and_var and predict_density for MC likelihood * factor out MC sampling * add comment for variance bias * add tests * fixes * use same integration as for GH quadrature in MonteCarloLikelihood.predict_mean_and_var() * . * increase rtol * move to proper use of super() * move MC integration to quadrature module, similar to ndiagquad * seed to make test deterministic * add Assert for shape of Y * tidy up studentT likelihood * fix for heteroskedastic likelihoods -- requires logp to always call the Y argument Y * fix doc * add assert and equivalence tests for SoftMax * remove erroneously added file * rename "probit" to inv_probit (which is what it actually is) * add assert for num_classes to SoftMax * fix whitespace * Update RELEASE.md * Update RELEASE.md | 27 June 2018, 11:19:26 UTC |
1b0f4b0 | ByungJun Lee | 26 June 2018, 10:18:10 UTC | Minibatch shape is unknow by default (#704) | 26 June 2018, 10:18:10 UTC |
9d23839 | st-- | 20 June 2018, 14:59:06 UTC | Notebook to demonstrate advanced usage such as combining GP with Neural Network (#712) * added notebook * added vanilla conv net accuracy * Tidy up notebook * ran notebook * ran notebook #2 * add untested notebooks to test_notebooks * fix bug in natural_gradients notebook * add nb_niter and nb_range to only run a small number of iterations in continuous integration notebook tests * make advanced_usage.ipynb use nb_iter() * don't add natural_gradients.ipynb and upper_bound.ipynb notebook tests in this PR * make advanced_usage run faster in CI * clean up advanced_usage * Update test_notebooks.py * fix notebook_niter * missing input_dim for CI * update notebook * another reset_default_graph_and_session call | 20 June 2018, 14:59:06 UTC |
3a16991 | Hugh Salimbeni | 19 June 2018, 15:00:11 UTC | Removed jitter in nat grads (#768) | 19 June 2018, 15:00:11 UTC |
61088fd | st-- | 19 June 2018, 13:19:35 UTC | Speed up notebooks (#789) Notebooks are slow to execute. Additional utility functions help to detect where notebook are run and control number of iterations (optimization, standard loops and cetera), therefore minimizing spent time on running notebook integration tests. | 19 June 2018, 13:19:35 UTC |
bb08f22 | Vincent Dutordoir | 18 June 2018, 17:04:06 UTC | Multi-output conditionals (#724) * Introduction of MultiOutputFeatures (Mof) and MultiOutputKernels (Mok). These are used to specify a particular setup of multi-output correlation. * Multiple-dispatch for conditional. This allows GPflow to select the most efficient conditional code depending on your choice of Mof and Mok. * Multiple-dispatch for Kuu and Kuf. Previously Kuu(.) and Kuf(.) were member functions of the feature class. This became cumbersome as the calculation of Kuu and Kuf also depends on the kernel used. In line with conditional we now also use multiple-dispatch to calculate Kuu and Kuf for a particular combination of Mok and Mof. * The actual maths to efficiently calculate the output-correlated conditional (credits to @markvdw ) * sample_conditional function that makes sure that the most efficient code is used to get a sample from the conditional distribution. * Minor: we updated a couple of models to use the new multi-output conditional. | 18 June 2018, 17:04:06 UTC |
6baeb43 | jch5f | 15 June 2018, 10:31:27 UTC | Likelihood/students t variance scaling (#777) * Add scaling to studentT conditional variance The conditional variance of the Student’s T distributions is proportional to the square of the scale of the distribution. See https://en.wikipedia.org/wiki/Student%27s_t-distribution#In_terms_of_sca ling_parameter_σ,_or_σ2. I’ve incorporated the correct scaling factor. * explicit scale dtype and tensor broadcasting Added an explicit data type for the Student’s T scale parameter, and made the broadcasting in the conditional_variance method explicit. | 15 June 2018, 10:31:27 UTC |
916458e | st-- | 13 June 2018, 13:21:24 UTC | ndiagquad logspace feature for improved numerical stability (#747) | 13 June 2018, 13:21:24 UTC |
8fa0d57 | Vincent Dutordoir | 13 June 2018, 09:39:00 UTC | Fix failing unittest under TensorFlow 1.8 (#788) | 13 June 2018, 09:39:00 UTC |