PyVBMC 1.0.1__tar.gz → 1.0.3__tar.gz

pyvbmc-1.0.3/CITATION.cff

@@ -0,0 +1,55 @@
+cff-version: "1.2.0"
+authors:
+- family-names: Huggins
+  given-names: Bobby
+  orcid: "https://orcid.org/0009-0006-3475-5964"
+- family-names: Li
+  given-names: Chengkun
+  orcid: "https://orcid.org/0000-0001-5848-910X"
+- family-names: Tobaben
+  given-names: Marlon
+  orcid: "https://orcid.org/0000-0002-9778-0853"
+- family-names: Aarnos
+  given-names: Mikko J.
+- family-names: Acerbi
+  given-names: Luigi
+  orcid: "https://orcid.org/0000-0001-7471-7336"
+contact:
+- family-names: Huggins
+  given-names: Bobby
+  orcid: "https://orcid.org/0009-0006-3475-5964"
+- family-names: Acerbi
+  given-names: Luigi
+  orcid: "https://orcid.org/0000-0001-7471-7336"
+doi: 10.5281/zenodo.7966315
+message: If you use this software, please cite our article in the
+  Journal of Open Source Software.
+preferred-citation:
+  authors:
+  - family-names: Huggins
+    given-names: Bobby
+    orcid: "https://orcid.org/0009-0006-3475-5964"
+  - family-names: Li
+    given-names: Chengkun
+    orcid: "https://orcid.org/0000-0001-5848-910X"
+  - family-names: Tobaben
+    given-names: Marlon
+    orcid: "https://orcid.org/0000-0002-9778-0853"
+  - family-names: Aarnos
+    given-names: Mikko J.
+  - family-names: Acerbi
+    given-names: Luigi
+    orcid: "https://orcid.org/0000-0001-7471-7336"
+  date-published: 2023-06-21
+  doi: 10.21105/joss.05428
+  issn: 2475-9066
+  issue: 86
+  journal: Journal of Open Source Software
+  publisher:
+    name: Open Journals
+  start: 5428
+  title: "PyVBMC: Efficient Bayesian inference in Python"
+  type: article
+  url: "https://joss.theoj.org/papers/10.21105/joss.05428"
+  volume: 8
+title: "PyVBMC: Efficient Bayesian inference in Python"

{PyVBMC-1.0.1 → pyvbmc-1.0.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: PyVBMC
-Version: 1.0.1
+Version: 1.0.3
 Summary: Variational Bayesian Monte Carlo in Python.
 License: BSD 3-Clause License
         
@@ -34,8 +34,28 @@ License: BSD 3-Clause License
         
 Requires-Python: >=3.9
 Description-Content-Type: text/markdown
-Provides-Extra: dev
 License-File: LICENSE
+Requires-Dist: cma>=3.1.0
+Requires-Dist: corner>=2.2.1
+Requires-Dist: dill>=0.3.5.1
+Requires-Dist: gpyreg>=0.1.0
+Requires-Dist: imageio>=2.13.5
+Requires-Dist: matplotlib>=3.9.0
+Requires-Dist: numpy<2.0.0
+Requires-Dist: plotly>=5.11.0
+Requires-Dist: pytest>=6.2.5
+Requires-Dist: pytest-mock>=3.6.1
+Requires-Dist: pytest-rerunfailures>=10.2
+Requires-Dist: scipy>=1.7.3
+Provides-Extra: dev
+Requires-Dist: myst_nb>=0.13.1; extra == "dev"
+Requires-Dist: numpydoc>=1.2.1; extra == "dev"
+Requires-Dist: pylint>=2.15.5; extra == "dev"
+Requires-Dist: pytest-cov>=4.0.0; extra == "dev"
+Requires-Dist: sphinx>=4.3.2; extra == "dev"
+Requires-Dist: sphinx-book-theme>=0.2.0; extra == "dev"
+Requires-Dist: build>=0.9.0; extra == "dev"
+Requires-Dist: black>=21.12.0; extra == "dev"
 
 ![PyVBMC](https://user-images.githubusercontent.com/70731267/225583733-6cc22c33-a198-4868-90be-3818206d1398.svg)
 
@@ -54,7 +74,7 @@ PyVBMC is a Python implementation of the Variational Bayesian Monte Carlo (VBMC)
 - an approximate posterior distribution of the model parameters;
 - an approximation — technically, an approximate lower bound — of the log model evidence (also known as log marginal likelihood or log Bayes factor), a metric used for [Bayesian model selection](https://en.wikipedia.org/wiki/Bayes_factor).
 
-Extensive benchmarks on both artificial test problems and a large number of real model-fitting problems from computational and cognitive neuroscience show that VBMC generally — and often vastly — outperforms alternative methods for sample-efficient Bayesian inference [[1,2](#references-and-citation)].
+Extensive benchmarks on both artificial test problems and a large number of real model-fitting problems from computational and cognitive neuroscience show that VBMC generally — and often vastly — outperforms alternative methods for sample-efficient Bayesian inference [[2,3](#references-and-citation)].
 
 ### Documentation
 
@@ -137,7 +157,7 @@ The `vp` object can be manipulated in various ways. For example, we can draw sam
 The quick start example above works for deterministic (noiseless) evaluations of the target log-density. Py(VBMC) also supports *noisy* evaluations of the target.
 Noisy evaluations often arise from simulation-based models, for which a direct expression of the (log) likelihood is not available.
 
-For information on how to run PyVBMC on a noisy target, see [this example notebook](examples/pyvbmc_example_5_noisy_likelihoods.ipynb) and the [VBMC FAQ](https://github.com/acerbilab/vbmc/wiki#noisy-target-function) (for MATLAB, but most concepts still apply).
+For information on how to run PyVBMC on a noisy target, see [this example notebook](examples/pyvbmc_example_6_noisy_likelihoods.ipynb) and the [VBMC FAQ](https://github.com/acerbilab/vbmc/wiki#noisy-target-function) (for MATLAB, but most concepts still apply).
 
 ## Next steps
 
@@ -161,7 +181,7 @@ In each iteration, PyVBMC uses *active sampling* to select which points to evalu
 
 In the figure above, we show an example PyVBMC run on a [Rosenbrock "banana" function](https://en.wikipedia.org/wiki/Rosenbrock_function). The bottom-left panel shows PyVBMC at work: in grayscale are samples from the variational posterior (drawn as small points) and the corresponding estimated density (drawn as contours). The solid orange circles are the active sampling points chosen at each iteration, and the hollow blue circles are the previously sampled points. The topmost and rightnmost panels show histograms of the marginal densities along the $x_1$ and $x_2$ dimensions, respectively. PyVBMC converges to an excellent approximation of the true posterior with a few dozens evaluations of the target density.
 
-See the VBMC papers [[1,2](#references-and-citation)] for more details.
+See the VBMC papers [[1-3](#references-and-citation)] for more details.
 
 ## Troubleshooting and contact
 
@@ -174,7 +194,7 @@ If you have trouble doing something with PyVBMC, spot bugs or strange behavior,
 
 ## References and citation
 
-1. Huggins, B., Li, C., Tobaben, M., Aarnos, M., & Acerbi, L. (2023). *PyVBMC: Efficient Bayesian inference in Python*. arXiv. https://arxiv.org/abs/2303.09519
+1. Huggins, B., Li, C., Tobaben, M., Aarnos, M., & Acerbi, L. (2023). [PyVBMC: Efficient Bayesian inference in Python](https://joss.theoj.org/papers/10.21105/joss.05428). *Journal of Open Source Software* 8(86), 5428, https://doi.org/10.21105/joss.05428.
 2. Acerbi, L. (2018). Variational Bayesian Monte Carlo. In *Advances in Neural Information Processing Systems 31*: 8222-8232. ([paper + supplement on arXiv](https://arxiv.org/abs/1810.05558), [NeurIPS Proceedings](https://papers.nips.cc/paper/8043-variational-bayesian-monte-carlo))
 3. Acerbi, L. (2020). Variational Bayesian Monte Carlo with Noisy Likelihoods. In *Advances in Neural Information Processing Systems 33*: 8211-8222 ([paper + supplement on arXiv](https://arxiv.org/abs/2006.08655), [NeurIPS Proceedings](https://papers.nips.cc/paper/2020/hash/5d40954183d62a82257835477ccad3d2-Abstract.html)).
 
@@ -199,14 +219,18 @@ You may also want to check out [Bayesian Adaptive Direct Search in Python (PyBAD
 
 ```BibTeX
 @article{huggins2023pyvbmc,
-  title = {PyVBMC: Efficient Bayesian inference in Python},
-  author = {Huggins, Bobby and Li, Chengkun and Tobaben, Marlon and Aarnos, Mikko J. and Acerbi, Luigi},
-  publisher = {preprint},
-  journal = {{arXiv}},
-  url = {https://arxiv.org/abs/2303.09519},
-  doi = {10.48550/ARXIV.2303.09519},
-  year = {2023},
-}
+    title = {PyVBMC: Efficient Bayesian inference in Python},
+    author = {Bobby Huggins and Chengkun Li and Marlon Tobaben and Mikko J. Aarnos and Luigi Acerbi},
+    publisher = {The Open Journal},
+    journal = {Journal of Open Source Software},
+    url = {https://doi.org/10.21105/joss.05428},
+    doi = {10.21105/joss.05428},
+    year = {2023},
+    volume = {8},
+    number = {86},
+    pages = {5428}
+  }
+
 @article{acerbi2018variational,
   title={{V}ariational {B}ayesian {M}onte {C}arlo},
   author={Acerbi, Luigi},

{PyVBMC-1.0.1 → pyvbmc-1.0.3}/PyVBMC.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: PyVBMC
-Version: 1.0.1
+Version: 1.0.3
 Summary: Variational Bayesian Monte Carlo in Python.
 License: BSD 3-Clause License
         
@@ -34,8 +34,28 @@ License: BSD 3-Clause License
         
 Requires-Python: >=3.9
 Description-Content-Type: text/markdown
-Provides-Extra: dev
 License-File: LICENSE
+Requires-Dist: cma>=3.1.0
+Requires-Dist: corner>=2.2.1
+Requires-Dist: dill>=0.3.5.1
+Requires-Dist: gpyreg>=0.1.0
+Requires-Dist: imageio>=2.13.5
+Requires-Dist: matplotlib>=3.9.0
+Requires-Dist: numpy<2.0.0
+Requires-Dist: plotly>=5.11.0
+Requires-Dist: pytest>=6.2.5
+Requires-Dist: pytest-mock>=3.6.1
+Requires-Dist: pytest-rerunfailures>=10.2
+Requires-Dist: scipy>=1.7.3
+Provides-Extra: dev
+Requires-Dist: myst_nb>=0.13.1; extra == "dev"
+Requires-Dist: numpydoc>=1.2.1; extra == "dev"
+Requires-Dist: pylint>=2.15.5; extra == "dev"
+Requires-Dist: pytest-cov>=4.0.0; extra == "dev"
+Requires-Dist: sphinx>=4.3.2; extra == "dev"
+Requires-Dist: sphinx-book-theme>=0.2.0; extra == "dev"
+Requires-Dist: build>=0.9.0; extra == "dev"
+Requires-Dist: black>=21.12.0; extra == "dev"
 
 ![PyVBMC](https://user-images.githubusercontent.com/70731267/225583733-6cc22c33-a198-4868-90be-3818206d1398.svg)
 
@@ -54,7 +74,7 @@ PyVBMC is a Python implementation of the Variational Bayesian Monte Carlo (VBMC)
 - an approximate posterior distribution of the model parameters;
 - an approximation — technically, an approximate lower bound — of the log model evidence (also known as log marginal likelihood or log Bayes factor), a metric used for [Bayesian model selection](https://en.wikipedia.org/wiki/Bayes_factor).
 
-Extensive benchmarks on both artificial test problems and a large number of real model-fitting problems from computational and cognitive neuroscience show that VBMC generally — and often vastly — outperforms alternative methods for sample-efficient Bayesian inference [[1,2](#references-and-citation)].
+Extensive benchmarks on both artificial test problems and a large number of real model-fitting problems from computational and cognitive neuroscience show that VBMC generally — and often vastly — outperforms alternative methods for sample-efficient Bayesian inference [[2,3](#references-and-citation)].
 
 ### Documentation
 
@@ -137,7 +157,7 @@ The `vp` object can be manipulated in various ways. For example, we can draw sam
 The quick start example above works for deterministic (noiseless) evaluations of the target log-density. Py(VBMC) also supports *noisy* evaluations of the target.
 Noisy evaluations often arise from simulation-based models, for which a direct expression of the (log) likelihood is not available.
 
-For information on how to run PyVBMC on a noisy target, see [this example notebook](examples/pyvbmc_example_5_noisy_likelihoods.ipynb) and the [VBMC FAQ](https://github.com/acerbilab/vbmc/wiki#noisy-target-function) (for MATLAB, but most concepts still apply).
+For information on how to run PyVBMC on a noisy target, see [this example notebook](examples/pyvbmc_example_6_noisy_likelihoods.ipynb) and the [VBMC FAQ](https://github.com/acerbilab/vbmc/wiki#noisy-target-function) (for MATLAB, but most concepts still apply).
 
 ## Next steps
 
@@ -161,7 +181,7 @@ In each iteration, PyVBMC uses *active sampling* to select which points to evalu
 
 In the figure above, we show an example PyVBMC run on a [Rosenbrock "banana" function](https://en.wikipedia.org/wiki/Rosenbrock_function). The bottom-left panel shows PyVBMC at work: in grayscale are samples from the variational posterior (drawn as small points) and the corresponding estimated density (drawn as contours). The solid orange circles are the active sampling points chosen at each iteration, and the hollow blue circles are the previously sampled points. The topmost and rightnmost panels show histograms of the marginal densities along the $x_1$ and $x_2$ dimensions, respectively. PyVBMC converges to an excellent approximation of the true posterior with a few dozens evaluations of the target density.
 
-See the VBMC papers [[1,2](#references-and-citation)] for more details.
+See the VBMC papers [[1-3](#references-and-citation)] for more details.
 
 ## Troubleshooting and contact
 
@@ -174,7 +194,7 @@ If you have trouble doing something with PyVBMC, spot bugs or strange behavior,
 
 ## References and citation
 
-1. Huggins, B., Li, C., Tobaben, M., Aarnos, M., & Acerbi, L. (2023). *PyVBMC: Efficient Bayesian inference in Python*. arXiv. https://arxiv.org/abs/2303.09519
+1. Huggins, B., Li, C., Tobaben, M., Aarnos, M., & Acerbi, L. (2023). [PyVBMC: Efficient Bayesian inference in Python](https://joss.theoj.org/papers/10.21105/joss.05428). *Journal of Open Source Software* 8(86), 5428, https://doi.org/10.21105/joss.05428.
 2. Acerbi, L. (2018). Variational Bayesian Monte Carlo. In *Advances in Neural Information Processing Systems 31*: 8222-8232. ([paper + supplement on arXiv](https://arxiv.org/abs/1810.05558), [NeurIPS Proceedings](https://papers.nips.cc/paper/8043-variational-bayesian-monte-carlo))
 3. Acerbi, L. (2020). Variational Bayesian Monte Carlo with Noisy Likelihoods. In *Advances in Neural Information Processing Systems 33*: 8211-8222 ([paper + supplement on arXiv](https://arxiv.org/abs/2006.08655), [NeurIPS Proceedings](https://papers.nips.cc/paper/2020/hash/5d40954183d62a82257835477ccad3d2-Abstract.html)).
 
@@ -199,14 +219,18 @@ You may also want to check out [Bayesian Adaptive Direct Search in Python (PyBAD
 
 ```BibTeX
 @article{huggins2023pyvbmc,
-  title = {PyVBMC: Efficient Bayesian inference in Python},
-  author = {Huggins, Bobby and Li, Chengkun and Tobaben, Marlon and Aarnos, Mikko J. and Acerbi, Luigi},
-  publisher = {preprint},
-  journal = {{arXiv}},
-  url = {https://arxiv.org/abs/2303.09519},
-  doi = {10.48550/ARXIV.2303.09519},
-  year = {2023},
-}
+    title = {PyVBMC: Efficient Bayesian inference in Python},
+    author = {Bobby Huggins and Chengkun Li and Marlon Tobaben and Mikko J. Aarnos and Luigi Acerbi},
+    publisher = {The Open Journal},
+    journal = {Journal of Open Source Software},
+    url = {https://doi.org/10.21105/joss.05428},
+    doi = {10.21105/joss.05428},
+    year = {2023},
+    volume = {8},
+    number = {86},
+    pages = {5428}
+  }
+
 @article{acerbi2018variational,
   title={{V}ariational {B}ayesian {M}onte {C}arlo},
   author={Acerbi, Luigi},

{PyVBMC-1.0.1 → pyvbmc-1.0.3}/PyVBMC.egg-info/SOURCES.txt

@@ -3,6 +3,7 @@
 .gitattributes
 .gitignore
 .pre-commit-config.yaml
+CITATION.cff
 LICENSE
 MANIFEST.in
 README.md

{PyVBMC-1.0.1 → pyvbmc-1.0.3}/PyVBMC.egg-info/requires.txt

@@ -3,8 +3,8 @@ corner>=2.2.1
 dill>=0.3.5.1
 gpyreg>=0.1.0
 imageio>=2.13.5
-matplotlib>=3.5.1
-numpy>=1.22.1
+matplotlib>=3.9.0
+numpy<2.0.0
 plotly>=5.11.0
 pytest>=6.2.5
 pytest-mock>=3.6.1

{PyVBMC-1.0.1 → pyvbmc-1.0.3}/README.md

@@ -15,7 +15,7 @@ PyVBMC is a Python implementation of the Variational Bayesian Monte Carlo (VBMC)
 - an approximate posterior distribution of the model parameters;
 - an approximation — technically, an approximate lower bound — of the log model evidence (also known as log marginal likelihood or log Bayes factor), a metric used for [Bayesian model selection](https://en.wikipedia.org/wiki/Bayes_factor).
 
-Extensive benchmarks on both artificial test problems and a large number of real model-fitting problems from computational and cognitive neuroscience show that VBMC generally — and often vastly — outperforms alternative methods for sample-efficient Bayesian inference [[1,2](#references-and-citation)].
+Extensive benchmarks on both artificial test problems and a large number of real model-fitting problems from computational and cognitive neuroscience show that VBMC generally — and often vastly — outperforms alternative methods for sample-efficient Bayesian inference [[2,3](#references-and-citation)].
 
 ### Documentation
 
@@ -98,7 +98,7 @@ The `vp` object can be manipulated in various ways. For example, we can draw sam
 The quick start example above works for deterministic (noiseless) evaluations of the target log-density. Py(VBMC) also supports *noisy* evaluations of the target.
 Noisy evaluations often arise from simulation-based models, for which a direct expression of the (log) likelihood is not available.
 
-For information on how to run PyVBMC on a noisy target, see [this example notebook](examples/pyvbmc_example_5_noisy_likelihoods.ipynb) and the [VBMC FAQ](https://github.com/acerbilab/vbmc/wiki#noisy-target-function) (for MATLAB, but most concepts still apply).
+For information on how to run PyVBMC on a noisy target, see [this example notebook](examples/pyvbmc_example_6_noisy_likelihoods.ipynb) and the [VBMC FAQ](https://github.com/acerbilab/vbmc/wiki#noisy-target-function) (for MATLAB, but most concepts still apply).
 
 ## Next steps
 
@@ -122,7 +122,7 @@ In each iteration, PyVBMC uses *active sampling* to select which points to evalu
 
 In the figure above, we show an example PyVBMC run on a [Rosenbrock "banana" function](https://en.wikipedia.org/wiki/Rosenbrock_function). The bottom-left panel shows PyVBMC at work: in grayscale are samples from the variational posterior (drawn as small points) and the corresponding estimated density (drawn as contours). The solid orange circles are the active sampling points chosen at each iteration, and the hollow blue circles are the previously sampled points. The topmost and rightnmost panels show histograms of the marginal densities along the $x_1$ and $x_2$ dimensions, respectively. PyVBMC converges to an excellent approximation of the true posterior with a few dozens evaluations of the target density.
 
-See the VBMC papers [[1,2](#references-and-citation)] for more details.
+See the VBMC papers [[1-3](#references-and-citation)] for more details.
 
 ## Troubleshooting and contact
 
@@ -135,7 +135,7 @@ If you have trouble doing something with PyVBMC, spot bugs or strange behavior,
 
 ## References and citation
 
-1. Huggins, B., Li, C., Tobaben, M., Aarnos, M., & Acerbi, L. (2023). *PyVBMC: Efficient Bayesian inference in Python*. arXiv. https://arxiv.org/abs/2303.09519
+1. Huggins, B., Li, C., Tobaben, M., Aarnos, M., & Acerbi, L. (2023). [PyVBMC: Efficient Bayesian inference in Python](https://joss.theoj.org/papers/10.21105/joss.05428). *Journal of Open Source Software* 8(86), 5428, https://doi.org/10.21105/joss.05428.
 2. Acerbi, L. (2018). Variational Bayesian Monte Carlo. In *Advances in Neural Information Processing Systems 31*: 8222-8232. ([paper + supplement on arXiv](https://arxiv.org/abs/1810.05558), [NeurIPS Proceedings](https://papers.nips.cc/paper/8043-variational-bayesian-monte-carlo))
 3. Acerbi, L. (2020). Variational Bayesian Monte Carlo with Noisy Likelihoods. In *Advances in Neural Information Processing Systems 33*: 8211-8222 ([paper + supplement on arXiv](https://arxiv.org/abs/2006.08655), [NeurIPS Proceedings](https://papers.nips.cc/paper/2020/hash/5d40954183d62a82257835477ccad3d2-Abstract.html)).
 
@@ -160,14 +160,18 @@ You may also want to check out [Bayesian Adaptive Direct Search in Python (PyBAD
 
 ```BibTeX
 @article{huggins2023pyvbmc,
-  title = {PyVBMC: Efficient Bayesian inference in Python},
-  author = {Huggins, Bobby and Li, Chengkun and Tobaben, Marlon and Aarnos, Mikko J. and Acerbi, Luigi},
-  publisher = {preprint},
-  journal = {{arXiv}},
-  url = {https://arxiv.org/abs/2303.09519},
-  doi = {10.48550/ARXIV.2303.09519},
-  year = {2023},
-}
+    title = {PyVBMC: Efficient Bayesian inference in Python},
+    author = {Bobby Huggins and Chengkun Li and Marlon Tobaben and Mikko J. Aarnos and Luigi Acerbi},
+    publisher = {The Open Journal},
+    journal = {Journal of Open Source Software},
+    url = {https://doi.org/10.21105/joss.05428},
+    doi = {10.21105/joss.05428},
+    year = {2023},
+    volume = {8},
+    number = {86},
+    pages = {5428}
+  }
+
 @article{acerbi2018variational,
   title={{V}ariational {B}ayesian {M}onte {C}arlo},
   author={Acerbi, Luigi},

{PyVBMC-1.0.1 → pyvbmc-1.0.3}/pyproject.toml

@@ -2,7 +2,7 @@
 
 [project]
 name = "PyVBMC"
-dynamic = ["version"]  # use git tags for version, via setuptools_scm
+dynamic = ["version"] # use git tags for version, via setuptools_scm
 description = "Variational Bayesian Monte Carlo in Python."
 readme = "README.md"
 license = { file = "LICENSE" }
@@ -12,8 +12,8 @@ dependencies = [
     "dill >= 0.3.5.1",
     "gpyreg >= 0.1.0",
     "imageio >= 2.13.5",
-    "matplotlib >= 3.5.1",
-    "numpy >= 1.22.1",
+    "matplotlib >= 3.9.0",
+    "numpy < 2.0.0",
     "plotly >= 5.11.0",
     "pytest >= 6.2.5",
     "pytest-mock >= 3.6.1",
@@ -26,7 +26,7 @@ requires-python = ">=3.9"
 include-package-data = true
 # Include examples in binary/wheel distribution:
 packages = ["pyvbmc", "pyvbmc.examples"]
-package-dir = {"pyvbmc.examples" = "examples"}
+package-dir = { "pyvbmc.examples" = "examples" }
 
 [tool.setuptools.package-data]
 # Make sure to include example Notebooks:
@@ -45,10 +45,7 @@ dev = [
 ]
 
 [build-system]
-requires = [
-    "setuptools >= 45",
-    "setuptools_scm[toml] >= 6.2",
-]
+requires = ["setuptools >= 45", "setuptools_scm[toml] >= 6.2"]
 build-backend = "setuptools.build_meta"
 
 [tool.black]

{PyVBMC-1.0.1 → pyvbmc-1.0.3}/pyvbmc/acquisition_functions/abstract_acq_fcn.py

@@ -135,7 +135,7 @@ class AbstractAcqFcn(ABC):
             np.any(X_orig < optim_state.get("lb_eps_orig"), axis=1),
             np.any(X_orig > optim_state.get("ub_eps_orig"), axis=1),
         )
-        acq[idx_bounds] = np.Inf
+        acq[idx_bounds] = np.inf
 
         # Re-shape to 1-D, if necessary (to avoid errors in cma.fmin)
         if acq.ndim > 1:

{PyVBMC-1.0.1 → pyvbmc-1.0.3}/pyvbmc/function_logger/function_logger.py

@@ -367,9 +367,9 @@ class FunctionLogger:
         ValueError
             Raise if there is more than one match for a duplicate entry.
         """
-        duplicate_flag = self.X == x
-        if np.any(duplicate_flag.all(axis=1)):
-            if np.sum(duplicate_flag.all(axis=1)) > 1:
+        duplicate_flag = (self.X == x).all(axis=1)
+        if np.any(duplicate_flag):
+            if np.sum(duplicate_flag) > 1:
                 raise ValueError("More than one match for duplicate entry.")
             idx = np.argwhere(duplicate_flag)[0, 0]
             N = self.n_evals[idx]

{PyVBMC-1.0.1 → pyvbmc-1.0.3}/pyvbmc/parameter_transformer/parameter_transformer.py

@@ -127,9 +127,9 @@ class ParameterTransformer:
 
             # Center in transformed space
             for i in range(D):
-                if np.isfinite(plb_tran[:, i]) and np.isfinite(pub_tran[:, i]):
-                    self.mu[i] = 0.5 * (plb_tran[:, i] + pub_tran[:, i])
-                    self.delta[i] = pub_tran[:, i] - plb_tran[:, i]
+                if np.isfinite(plb_tran[0, i]) and np.isfinite(pub_tran[0, i]):
+                    self.mu[i] = 0.5 * (plb_tran[0, i] + pub_tran[0, i])
+                    self.delta[i] = pub_tran[0, i] - plb_tran[0, i]
 
     @handle_0D_1D_input(patched_kwargs=["x"], patched_argpos=[0])
     def __call__(self, x: np.ndarray):
@@ -148,7 +148,7 @@ class ParameterTransformer:
         u : np.ndarray
             The variables transformed to unconstrained variables.
         """
-
+        x = x.astype(float)
         u = np.copy(x)
 
         # Unbounded scalars (possibly center and rescale)
@@ -160,7 +160,9 @@ class ParameterTransformer:
         for t in self.bounded_types:
             mask = self.type == t
             if np.any(mask):
-                u[:, mask] = self._bounded_transforms[t]["direct"](x, mask)
+                u[:, mask] = self._bounded_transforms[t]["direct"](
+                    self, x, mask
+                )
 
         # Rotoscale whitening:
         # Rotate and rescale points in transformed space.
@@ -187,7 +189,7 @@ class ParameterTransformer:
         x : np.ndarray
             The original variables which result of the transformation.
         """
-
+        u = u.astype(float)
         x = np.copy(u)
 
         # Rotoscale whitening:
@@ -208,7 +210,9 @@ class ParameterTransformer:
         for t in self.bounded_types:
             mask = self.type == t
             if np.any(mask):
-                xNew[:, mask] = self._bounded_transforms[t]["inverse"](x, mask)
+                xNew[:, mask] = self._bounded_transforms[t]["inverse"](
+                    self, x, mask
+                )
 
         return xNew
 
@@ -232,6 +236,7 @@ class ParameterTransformer:
         p : np.ndarray
             The log absolute determinant of the Jacobian.
         """
+        u = u.astype(float)
         u_c = np.copy(u)
 
         # Rotoscale whitening:
@@ -252,7 +257,9 @@ class ParameterTransformer:
         for t in self.bounded_types:
             mask = self.type == t
             if np.any(mask):
-                p[:, mask] = self._bounded_transforms[t]["jacobian"](u_c, mask)
+                p[:, mask] = self._bounded_transforms[t]["jacobian"](
+                    self, u_c, mask
+                )
 
         # Whitening/rotoscaling density correction:
         if self.scale is not None:
@@ -274,7 +281,7 @@ class ParameterTransformer:
             if t == 3:
                 # logit
 
-                def bounded_transform(x, mask):
+                def bounded_transform(self, x, mask):
                     return _center(
                         _logit(
                             _to_unit_interval(
@@ -289,7 +296,7 @@ class ParameterTransformer:
 
                 self._bounded_transforms[t]["direct"] = bounded_transform
 
-                def bounded_inverse(u, mask):
+                def bounded_inverse(self, u, mask):
                     return _from_unit_interval(
                         _inverse_logit(
                             _uncenter(
@@ -302,7 +309,7 @@ class ParameterTransformer:
 
                 self._bounded_transforms[t]["inverse"] = bounded_inverse
 
-                def bounded_jacobian(u, mask):
+                def bounded_jacobian(self, u, mask):
                     j1 = np.log(self.ub_orig[:, mask] - self.lb_orig[:, mask])
                     y = _uncenter(u[:, mask], self.mu[mask], self.delta[mask])
                     z = -np.log1p(np.exp(-y))
@@ -315,7 +322,7 @@ class ParameterTransformer:
             elif t == 12:
                 # probit: inverse normal CDF (probit) transform (default)
 
-                def bounded_transform(x, mask):
+                def bounded_transform(self, x, mask):
                     return _center(
                         _probit(
                             _to_unit_interval(
@@ -330,7 +337,7 @@ class ParameterTransformer:
 
                 self._bounded_transforms[t]["direct"] = bounded_transform
 
-                def bounded_inverse(u, mask):
+                def bounded_inverse(self, u, mask):
                     return _from_unit_interval(
                         _inverse_probit(
                             _uncenter(
@@ -343,7 +350,7 @@ class ParameterTransformer:
 
                 self._bounded_transforms[t]["inverse"] = bounded_inverse
 
-                def bounded_jacobian(u, mask):
+                def bounded_jacobian(self, u, mask):
                     j1 = np.log(self.ub_orig[:, mask] - self.lb_orig[:, mask])
                     y = _uncenter(u[:, mask], self.mu[mask], self.delta[mask])
                     j2 = -0.5 * np.log(2 * np.pi) - 0.5 * y**2
@@ -355,7 +362,7 @@ class ParameterTransformer:
             elif t == 13:
                 # student4: Student's T with nu=4 CDF transform
 
-                def bounded_transform(x, mask):
+                def bounded_transform(self, x, mask):
                     return _center(
                         _student4(
                             _to_unit_interval(
@@ -370,7 +377,7 @@ class ParameterTransformer:
 
                 self._bounded_transforms[t]["direct"] = bounded_transform
 
-                def bounded_inverse(u, mask):
+                def bounded_inverse(self, u, mask):
                     return _from_unit_interval(
                         _inverse_student4(
                             _uncenter(
@@ -383,7 +390,7 @@ class ParameterTransformer:
 
                 self._bounded_transforms[t]["inverse"] = bounded_inverse
 
-                def bounded_jacobian(u, mask):
+                def bounded_jacobian(self, u, mask):
                     j1 = np.log(self.ub_orig[:, mask] - self.lb_orig[:, mask])
                     y = _uncenter(u[:, mask], self.mu[mask], self.delta[mask])
                     j2 = np.log(3 / 8) - (5 / 2) * np.log1p(y**2 / 4)
@@ -496,7 +503,7 @@ def _uncenter(v, mu, delta):
 
 def _logit(z):
     # prevent divide by zero
-    u = np.zeros(z.shape)
+    u = np.zeros_like(z)
     u[z == 0] = -np.inf
     u[z == 1] = np.inf
 
@@ -505,7 +512,12 @@ def _logit(z):
 
 
 def _inverse_logit(u):
-    return 1 / (1 + np.exp(-u))
+    # prevent overflow
+    z = np.zeros_like(u)
+    mask = -u > np.log(np.finfo(np.float64).max)
+    z[mask] = 0.0
+    z[~mask] = 1 / (1 + np.exp(-u[~mask]))
+    return z
 
 
 def _probit(z):
@@ -518,7 +530,11 @@ def _inverse_probit(u):
 
 def _student4(z):
     aa = np.sqrt(4 * z * (1 - z))
-    q = np.cos(np.arccos(aa) / 3) / aa
+    # prevent divide by zero
+    mask = aa == 0.0
+    q = np.zeros_like(z)
+    q[mask] = np.inf
+    q[~mask] = np.cos(np.arccos(aa[~mask]) / 3) / aa[~mask]
     return np.sign(z - 0.5) * (2 * np.sqrt(q - 1))
 
 

{PyVBMC-1.0.1 → pyvbmc-1.0.3}/pyvbmc/priors/prior.py

@@ -60,7 +60,7 @@ class Prior(ABC):
         return np.exp(self.log_pdf(x, keepdims=keepdims))
 
     def support(self):
-        """Returns the support of the distribution.
+        r"""Returns the support of the distribution.
 
         Used to test that the distribution integrates to one, so it is also
         acceptable to return a box which bounds the support of the

{PyVBMC-1.0.1 → pyvbmc-1.0.3}/pyvbmc/priors/spline_trapezoidal.py

@@ -7,7 +7,7 @@ from pyvbmc.priors import Prior, tile_inputs
 
 
 class SplineTrapezoidal(Prior):
-    """Multivariate spline-trapezoidal prior.
+    r"""Multivariate spline-trapezoidal prior.
 
     A prior distribution represented by a density with external bounds ``a``
     and ``b`` and internal points ``u`` and ``v``. Each marginal distribution

{PyVBMC-1.0.1 → pyvbmc-1.0.3}/pyvbmc/priors/trapezoidal.py

@@ -7,7 +7,7 @@ from pyvbmc.priors import Prior, tile_inputs
 
 
 class Trapezoidal(Prior):
-    """Multivariate trapezoidal prior.
+    r"""Multivariate trapezoidal prior.
 
     A prior distribution represented by a density with external bounds ``a``
     and ``b`` and internal points ``u`` and ``v``. Each marginal distribution

{PyVBMC-1.0.1 → pyvbmc-1.0.3}/pyvbmc/stats/get_hpd.py

@@ -40,6 +40,6 @@ def get_hpd(X: np.ndarray, y: np.ndarray, hpd_frac: float = 0.8):
     if hpd_N > 0:
         hpd_range = np.max(hpd_X, axis=0) - np.min(hpd_X, axis=0)
     else:
-        hpd_range = np.full((D), np.NaN)
+        hpd_range = np.full((D), np.nan)
 
     return hpd_X, hpd_y, hpd_range, indices

{PyVBMC-1.0.1 → pyvbmc-1.0.3}/pyvbmc/stats/kde_1d.py

@@ -42,8 +42,8 @@ def _fixed_point(
     Note that the factor of 2.0 in the definition of f is correct. See longer
     discussion here: https://github.com/tommyod/KDEpy/issues/95
     """
-    i_range_squared = np.asfarray(i_range_squared, dtype=np.float64)
-    a2 = np.asfarray(a2, dtype=np.float64)
+    i_range_squared = np.asarray(i_range_squared, dtype=np.float64)
+    a2 = np.asarray(a2, dtype=np.float64)
     ell = 7
     f = (
         2.0
@@ -59,7 +59,7 @@ def _fixed_point(
         return -1
 
     for s in reversed(range(2, ell)):
-        odd_numbers_prod = np.product(
+        odd_numbers_prod = np.prod(
             np.arange(1, 2 * s + 1, 2, dtype=np.float64)
         )
         K0 = odd_numbers_prod / np.sqrt(2.0 * np.pi)