Next meeting: April 30th, 2025 (Wednesday) @ 2pm UTC Join via Zoom at https://us06web.zoom.us/j/6345425936?pwd=aDVFQzVmbk9SVU5jU0Jwc0s3YWUrdz09If you need a passcode for zoom, scipy should work. Meeting notes archive at https://github.com/scipy/archive/tree/main/community_meetings Code of Conduct We want to take a moment to remind you that this meeting, like all project spaces is meant to be open, welcoming, diverse, inclusive, and it's important for us to have a healthy community. Like all SciPy spaces, and everyone participating in them, this meeting will follow our code of conduct. If you haven't read it yet, please take some time to do so later on as it already applies to you. For now, in short, please be kind and generous towards one another. Agenda for April 30th, 2025 please add your names (and github handle in parenthesis). This will make it easier to stay in touch later on issues and pull requests (PRs) 😉

Hi everyone, After years in the making, maintainers @mdhaber, @tupui, @ev-br, @chrisb83, @tirthasheshpatel, @steppi, and @dschmitz89 are happy to present the following RFC. RFC: stats: univariate distribution infrastructure This work is part of the CZI EOSS5-0000000176 grant (Matt and Pamphile). We reached out to the wider Scientific Python community using a variety of communication channels to share our CZI work.. The proposal was well received and we've taken the feedback we've received into consideration.

Effort to get scipy.stats._distribution_infrastructure into SciPy. Working PR: https://github.com/mdhaber/scipy/pull/110 Meeting 1: Hello! Presentation of the project (2024-01-31 20 CET) Agenda Cheers together, it’s been a while since we had a coffee talk :) Go around the new proposed API Raise concerns, draft solutions, distribute workload. Matt shared a notebook with a demo:

Alt title: ? Short Summary Uncertainties are everywhere! Whether you are developing a new AI system, running complex simulations or making an experiment in a lab, uncertainties influence the system. And you need a way to understand how these impact your results. SimDec offers a novel visual way to understand the intracate role that uncertainties play. Thanks to a clear API and our dashboard, we are making uncertainty analysis accessible to everyone. Abstract From real life experiments to numerical simulations, uncertainties play a crucial role in the system under study. With the avenment of Artificial Intelligence and new regulations such as the AI Act or the Better Regulation Guideline, there is a growing need for explainability and impact assessments of systems under uncertainties.

Letter of intention: https://hackmd.io/@tupui/sailb-czi/edit GDrive: https://drive.google.com/drive/folders/1QqGAVU-LGS5uaARKNk926TNYQXDJGyRx?usp=sharing Meetings 2023-11-07 (https://meet.google.com/ivk-ihit-bwe) Agenda Hi :wave: Our availability, work appetite and :heavy_dollar_sign:Takuya: Code, doc Will: DEI, supervision role, researcher involved for some tasks Pamphile: Code, involve people like Andrea

Or just Sensitivity Analysis Tools for Biomedical Research? On the portal for the submission, they ask to list projects that would be involved. I list SALib as the first project and SciPy as the second one. Hopefully that would help our case. This is why I also added some wording around SciPy. Note that due to the word limit, we would not put full citations, this would be done on the full proposal Proposal Summary/Scope of Work Provide a short summary of the work being proposed (maximum of 500 words) This proposal has 4 goals: (1) Develop a new coherent API, (2) Develop a framework for uncertainty visualization, (3) General maintenance, and (4) outreach.

Some review of the branch new_distribution[2] First pass You can update parameters with the update_parameters method. (e.g. dist.update_parameters(a=[1.0, 2.0]).As a user, if I can do dist.a, that could be surprising to not be able to do dist.a = .... plot: really cool to include that feature! I am not sure though about the list. I think I prefer composing by adding another layer to an axis. i.e. we would not allow passing a list and this simplifies the API. sample/qmc_sample: I know we have the distinction in UNU.RAN. But from a user perspective is this really needed? I would use a setting for that. And consider as well making QMC the default ;) This is a more "tricky" point OC. Should we make a separate folder for all that?

Agenda for ..., 2023 Present: Pamphile Roy, ... Feel free to add items for discussion to this agenda! [name=Pamphile] ...

General Rules 2FA for everything Limit access to what you need only Limit token validity and access Careful about SSH keys and env variables No sharing of credentials Define groups and backups Infrastructure Code

Welcome! You are now officially a SciPy maintainer. The goal of this document is to give you some information and more context that can help you navigate our waters with more confidence. How does SciPy fit-in A bit of history So many actors: NumFOCUS - Scientific Python - Tydelift - OpenCollective - CZI -PyData - SciPy conferences https://opencollective.com/scipy https://scientific-python.org https://numfocus.org/programs/small-development-grants

Alt title: More than error bars: Sensitivity Analysis in Python with scipy.stats.sobol_indices Short Summary The study of uncertainties can provide invaluable insights on a system or model. A central question is: what parameter of my system is the most important. This is sensitivity analysis and the Sobol' indices are helping answer this question. Thanks to new developments in SciPy, there is now a function for that: scipy.stats.sobol_indices. This method has been successfully used to analyse systems in various fields, enabling practicionners make critical decisions: Can I neglect this parameter? Is it more important than my other parameters? Can I give it a ranking? Can I explain my machine learning model's parameters?

Summary Statement of need NumPy random number generators (numpy.random) have become the de-facto standard for sampling random numbers in the scientific Python ecosystem. These methods are fast and reliable, and the results are repeatable when a seed is provided. As a foundational tool, NumPy only provides classical Monte Carlo (MC) methods. Sampling in high dimensions with MC produces a lot of gaps and clusters of points. When these random numbers are used in algorithms (including sampling, numerical integration, optimization) to solve deterministic problems, the resulting MC methods have a slow

This document Slides Scientific Python Communit Discord Project homepage SPECs Attending: Stéfan van der Walt @stefanv

Short Summary NumPy random number generators and SciPy distributions are widely used to get random numbers. However, challenges might arise in the following situations: (i) sampling from non-standard distributions can be slow if a custom implementation is not available and (ii) sampling in high dimensions leads to poor convergence rates. Thanks to new developments in SciPy, there is an answer to these problems with two new submodules: scipy.stats.sampling and scipy.stats.qmc. Abstract NumPy random number generators (numpy.random) and SciPy distributions (scipy.stats) have become the standard way of sampling random numbers in the scientific Python ecosystem. These methods are fast and reliable, and the results are repeatable when a random seed is provided. This talk addresses two challenges with these methods: It's difficult to sample from a new or non-standard distribution. For example, naive methods to numerically invert the cumulative distribution function can be too slow in practice, even for simple probability density functions.

Proposal Add a Sensitivity Analysis (SA) function. The function would compute Sobol' indices [1,2]. Consider a function f with parameters x1, x2 and x3. Hence y=f(x1,x2,x3). We are interested to know which parameter has the most impact, in terms of variance, on the value y. I believe scikit-learn, and the wider scientific community, would greatly benefit to have such tool. I believe scikit-learn has something related with feature_importances_ in some regressors. As an expert in the field, I would propose to develop the initial functionalities and provide long term support. Background

Jul 11-17, 2022 https://www.scipy2022.scipy.org/talk-poster-presentations Scientific Python - By maintainers and for maintainers Short Summary The future of Scientific Python is here. Behind the scenes, core scientific projects have started coordinating to better unify the community. Learn more about the goals of this unification effort, opportunities that exist, and, most importantly, how to join the adventure. Abstract The Scientific Python project's vision is to help pave the way towards a unified, expanded scientific Python community. It focuses its efforts along two primary axes: (i) to create a joint community around all scientific projects and (ii)_to support maintainers by building cross-cutting technical infrastructure and tools. This talk will focus on the second aspect, presenting current and future initiatives.

Organization Post-sprint For continued support and following up on your pull requests: Reach out to SciPy on their mailing-list or GitHub. Open an issue or comment on an issue. The Team SciPy Mentors (all Core Contributors) SciPy is a library that is used around the world. Open source sprints are typically limited to where contributors live or where major conferences are held. We are fortunate to have support from several SciPy contributors:

Time: 09:00 CEST Join via Zoom at https://zoom.us/j/93827717491?pwd=UGdkMzlnWEFjbDNQUmxpRkp6L0VLdz09 (or dial-in) Previous community meetings Present: Pamphile Roy, Tirth Patel, Bas Van Beek, Gemma Topics Code Specifics

Time: xx:00 CEST Join via Zoom at https://zoom.us/j/93827717491?pwd=UGdkMzlnWEFjbDNQUmxpRkp6L0VLdz09 (or dial-in) Previous community meetings Present: Tirth Patel, Bas van Beek, Melissa Mendonca, Ralf Gommers, Ross Barnowski Follow-up from last meeting / discussions On using community planning to manage GitHub issues/PR Optimization refactoring