{"id":22901308,"url":"https://github.com/mooerslab/jointprob1d","last_synced_at":"2026-03-19T23:57:13.178Z","repository":{"id":165234134,"uuid":"526986945","full_name":"MooersLab/jointprob1D","owner":"MooersLab","description":"Notes for the jointprob group","archived":false,"fork":false,"pushed_at":"2024-05-28T14:31:42.000Z","size":7183,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-04-01T05:41:59.996Z","etag":null,"topics":["bayesian-data-analysis","bayesian-inference","bayesian-statistics"],"latest_commit_sha":null,"homepage":"","language":"HTML","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/MooersLab.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2022-08-20T16:55:33.000Z","updated_at":"2024-05-28T14:31:45.000Z","dependencies_parsed_at":"2024-12-14T01:42:56.474Z","dependency_job_id":null,"html_url":"https://github.com/MooersLab/jointprob1D","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/MooersLab/jointprob1D","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MooersLab%2Fjointprob1D","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MooersLab%2Fjointprob1D/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MooersLab%2Fjointprob1D/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MooersLab%2Fjointprob1D/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/MooersLab","download_url":"https://codeload.github.com/MooersLab/jointprob1D/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MooersLab%2Fjointprob1D/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28336625,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-12T06:09:07.588Z","status":"ssl_error","status_checked_at":"2026-01-12T06:05:18.301Z","response_time":98,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["bayesian-data-analysis","bayesian-inference","bayesian-statistics"],"created_at":"2024-12-14T01:32:50.446Z","updated_at":"2026-01-12T07:40:13.166Z","avatar_url":"https://github.com/MooersLab.png","language":"HTML","funding_links":[],"categories":[],"sub_categories":[],"readme":"![version](https://img.shields.io/static/v1?label=jointProb1D\u0026message=0.2\u0026color=brightcolor)\n[![license: mit](https://img.shields.io/badge/license-mit-blue.svg)](https://opensource.org/licenses/mit)\n\n\n# jointprob\n\nThe [jointprob community](https://scicloj.github.io/docs/community/groups/jointprob/) is a Bayesian data analysis (BDA) study group.\nWe aimed to work through Richard McElreath's *Rethinking Statistics, 2nd Ed*.\nWe met for two hours once every two weeks.\nWe made it through chapter 7.\n \nThere were initially about 30 people interested in jointprob last summer.\nFour parallel sections were held to accommodate different schedules and time zones.\nI participated in Section D, which met on Saturdays.\nSection D was the last section left standing as of December.\nWe were down to 4-6 participants.\nIt was time for a refresh.\n\nThe consensus was to shift our focus to the book *Bayesian Modeling and Computation In Python (BMCP)* by Osvaldo Martin, Ravin Kumar, and Junpeng Lao.\nWe had an organizing meeting on January 7th.\n24 people attended. \n\nRavin Kumar also attended and spoke for about an hour.\nHe provided invaluable insights about how best to read his book.\nHe stressed that his book emphasizes the practice side before the theory.\nHe recommended McElreath's book for the theory.\nThe book uses three probabilistic programming languages: PyMC, tensorflow probability, and numpyro.\n\nRavin is a mechanical engineer who quit SpaceX to spend time teaching himself BDA.\nHe started contributing to the [PyMC3](https://www.pymc.io/welcome.html) project on GitHub.\nHe then got hired at Google to apply BDA to various problems.\nHis enthusiasm for BDA is contagious, as seen in the [video](https://www.youtube.com/watch?v=foSPfzYs4yY) that he made about Bayesian vs. Frequentist approaches.\n\nHe and others are offering a paid [course](https://www.intuitivebayes.com/introductorycourse) for professionals.\nThe introductory course is available and two more are preparation. \n\nThe [book](https://bayesiancomputationbook.com/welcome.html) is available online for free.\nThe [code](https://github.com/BayesianModelingandComputationInPython/BookCode_Edition1) for the book is located on GitHub.\nThe code in the book uses PyMC3, but the current version of pymc is version 5.\nThe code has not been translated to PyMC5.\nThe pymc community has a [discourse channel](https://discourse.pymc.io/) and an upcoming [webinar](https://pymcon.com/about) series. \nThey had a similar webinar [series](https://www.youtube.com/watch?v=UznM_-_760Y\u0026list=PLD1x-BW9UdeHN2vwR6kIApJATd2jZzeya\u0026index=1) in 2020. \n\nThe first four chapters are the heart of the book.\nThey are all that you need to start practicing BDA.\nChapters 5 - 8 are specialized topics that cover areas that most people will use.\nChapters 9, 10, and 11 contain appendix material.\n\n- 1 \u0026 2 Basics\n- 3 \u0026 4 Linear models and their applications\n- 5 Splines\n- 6 Time series\n- 7 BART\n- 8 ABC\n- 9 Bayesian workflow\n- 10 PPLs\n- 11 Appendical Topics (many theory topics are nicely summarized here )\n\nDaniel Slutsky leads the meetings.\nRyan Orsinger is the community organizer.\nThe SciCloj community sponsors the jointprob events.\nThis community is developing scientific computing tools in Clojure.\n\n\nThese are the guiding principles for the group (Thanks to Ryan Onsinger!):\n\n- *No experts.* We do not assume that anybody is an expert in the field. We come to learn together with a student mindset.\n\n- *A clear path.* We will be very thoughtful about the agenda and where we wish to go. We will continually rethink and adapt our pathway going there.\n\n- *Confused together.* It is just fine to be confused. We will be there together and seek clarity together.\n\n- *Being active.* We encourage members to learn independently and take on projects. In a sense, its purpose is (also) to support those individual journeys.\n\n- *Mutual curiosity.* We make serious efforts to be inclusive to participants of various backgrounds. The different perspectives of our friends are part of what we wish to learn.\n\nYou do not need to do the reading in advance, but you will get more out of the meetings if you do so.\nYou will also get more out of the meetings by presenting a portion of the reading: the best way to learn to try to teach the material.\nThis takes preparatory time. \nI found that 6-8 hours were required to assemble a 30-40-minute talk.\n\n\n## meeting1.Rmd\n\nThe Rmarkdown file that I presented in the first meeting of section D on Saturday, August 20, 2022. \nI covered Chapter 1 of McElreath and Chapters 1-4 of Grolemund and Wickham [*R for Data Science*](https://bookdown.org/roy_schumacher/r4ds/).\n\n## meeting2b.Rmd\n\nI edited this Rmarkdown file that I presented in the second meeting of section D on Saturday, September 3, 2022. \nDaniel Slutsky gave an excellent 80-minute presentation about computing the posterior distribution, which prepared me well to present how to use grid approximation to estimate the posterior distribution in R.\n\nI included the suggested exercises from Chapter 2 of McElreath's *Rethinking Statistics, 2nd Ed*. \nNext, I ventured off and tried a triangular distribution from the `extraDistr` package as a prior.\n\nI added some embellishments, such as normalizing the prior, summing the prior, and estimating the likelihood, as sanity checks.\nThese embellishments were not required to compute the correct posterior, but they deepened the understanding of what was happening.\n\n## meeting3b.Rmd\n\nI presented this Rmarkdown file in the third meeting of section D on Saturday, September 17, 2202.\nIt recaps the grid approximation presentation and then covers the two other *motors* of the Bayesian data analysis engine: quadratic approximation and Markov Chain Monte Carlo.\n\n\n## SectionD10.ipynb\n\nI presented this Jupyter notebook on information theory at the Christmas Eve meeting of jointprob.\nI included material from chapter 11 of BMCP.\nOne code cell does not work.\n\n\n## Appendix of Useful Links\n\n\n### Programs \n\n#### Stan\n\n[Stan](https://mc-stan.org/) implements that Hamiltonian Monte Carlo (HMC) with the No U-turn Sampler, which searches parameter space much faster than MCMC samplers.\nHMC cannot handle models with discrete parameters. These parameters have to be marginalized out via algebra. See the [Stan Users Guide](https://mc-stan.org/docs/stan-users-guide/latent-discrete.html).\n\n##### cmdstan\n\n[cmdstan](https://mc-stan.org/users/interfaces/cmdstan) is probably the best way to access the current version of Stan. \nPyStan and RStan lag behind by several versions.\nThere are cmdstanpy and cmdstanR to interface with cmdstan from Python or R.\n\n##### BridgeStan\n\n[BridgeStan](https://github.com/roualdes/bridgestan) is a new way to interact with Stan model objects from R, Python, Julia, Rust, or C.\nThey talk to each via their C interfaces.\nBrdigeStan allows you to access the methods of Stan model objects from a program than C++, which Stan is written in.\nYou can also use [ArviZ](https://www.arviz.org/en/latest/) to make plots from the sampled posterior for an Stan object.\n\n##### nutpie\n\n[nutpie](https://github.com/pymc-devs/nutpie) is a rust based interface to both Stan and PyMC.\nIt is on version 0.1 and very underdeveloped.\nThere is only a working example for modeling the mean of a sample from stan model.\n\n\n#### RStan (C++ wrapped in R)\n\n\n#### PyMC (Python)\n\n#####  Quick tutorial in PyMC4 \n\nPyMC3 from the earlier PeerJ paper was translated to [PyMC4](https://www.pymc.io/projects/docs/en/stable/learn/core_notebooks/pymc_overview.html#pymc-overview).\n\nNote that [PyMC](https://www.pymc.io/welcome.html) is now in the version 5 series.\nThe appending of a number has been dropped.\n\n#### Turing (Julia)\n\n\n\n#### Anglican (Clojure)\n\n### Books \n\nMany of the popular books on BDA have associated computer code.\nOften, this computer code has been translated into other programming languages by kind people.\n\n### Bayesian Analysis in Python (BAP)\n\nThe second edition of BAP's code in PyMC3.11 is available (https://github.com/aloctavodia/BAP)\n\n\n### Rethinking Statistics \n\n#### PyMC variation\nNote that McElreath's book has been fully translated into [PyMC3](https://github.com/pymc-devs/pymc-resources/tree/main/Rethinking_2) and largely translated into PyMC4, so the Rethinking Statistics book is ahead of the BMCP book in this regard.\n\n### Rethinking has been translated into Julia\nMcElreath's book has been translated into [Julia](https://github.com/StatisticalRethinkingJulia).\n\n### BMCP has been translated into Julia\n\n\n#### Julia and the Turing Package\n\n[Fun introduction](https://storopoli.github.io/Bayesian-Julia/)\n\n### John Krusche's Doing Bayesian Data Analysis (Puppydog book) in PyMC3\n\nhttps://github.com/JWarmenhoven/DBDA-python\n\n\n### Bayesian Ddata Analysis Edition 3 in PyMC3\n\nThis is a more advanced (aka harder to read) book that was published with a minimal amount of code for Stan.\nThe translation of the book is still a work in [progress](https://github.com/pymc-devs/pymc-resources/tree/main/BDA3).\n\n\n### Regression and Other Stories \n\n\n#### R code\nThis is a more accessible book. It is an update of an earlier book by Gelman and Hill. It is free and [on-line](https://statmodeling.stat.columbia.edu/2022/01/27/regression-and-other-stories-free-pdf/).\n\n#### PyMC3\n\n It is being translated into the bambi wrapper for [PyMC](https://github.com/bambinos/educational-resources). Nothing has happened in two years.\n\n\n## Update History\n\n|Version      | Changes                                         | Date            |\n|:-----------:|:-----------------------------------------------:|:---------------:|\n| Version 0.2 |  Fixed typos in README.md                       | 2024 April 10    |\n| Version 0.3 | Added more links.                               | 2024 May 28    |\n\n\n## Sources of funding\n\n- NIH: R01 CA242845\n- NIH: R01 AI088011\n- NIH: P30 CA225520 (PI: R. Mannel)\n- NIH P20GM103640 and P30GM145423 (PI: A. West)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmooerslab%2Fjointprob1d","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmooerslab%2Fjointprob1d","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmooerslab%2Fjointprob1d/lists"}