sbi developers
@sbi_devs
Followers
222
Following
77
Media
16
Statuses
46
Simulation-based inference (SBI) toolkit in PyTorch. Tweeted by @deismic_ and @janfiete
Joined March 2022
sbi 0.24.0 is out! 🎉 This comes with important new features:.- 🎯 Score-based i.i.d sampling.- 🔀 Simultaneous estimation of multiple discrete and continuous parameters or data. - 📊: mini-sbibm for quick benchmarking. Just in time for our 1-week SBI hackathon starting tomorrow!.
1
1
6
🙏 Please help us improve the SBI toolbox! 🙏 . In preparation for the SBI Hackathon, we’re running a user study to learn what we can improve and how we can grow. 👉 Please share your thoughts here: Your input will make a big difference—thank you! 🙌.
docs.google.com
Dear SBI User, We're committed to making the SBI toolbox the best possible tool for your simulation-based inference needs. To achieve this, we're seeking your feedback on the SBI toolbox. We are...
0
4
4
What to expect:.- Coding sessions to enhance the sbi toolbox.- Research talks & lightning talks.- Networking & idea exchange. In-person attendance is encouraged but a remote option is available. Free to attend, but seats are limited. Beginners welcome! 🤗.
0
0
1
🚀 Join the 4th SBI Hackathon! 🚀. The last hackathon was a fantastic milestone in forming a collaborative open-source community around SBI. Be part of it this year as we build on that momentum!. 📅 March 17–21, 2025.📍 Tübingen or remote.👉 Details: 🧵👇.
1
2
7
🙌 Huge thanks to our contributors for this release, including 5 first-time contributors! 🌟. Special shoutout to:.emmanuel-ferdman, CompiledAtBirth, tvwenger, matthewfeickert, and manuel-morales-a 🎉. Let us know what you think of the new version!.
0
0
1
✨ Highlights in v0.23.3: .- sbi is now available via @condaforge 🛠️ .- we now support MCMC sampling with multiple i.i.d. conditions 🎯 (this is for you, decision-making researchers). 💡 Plus, improved docs here and there, clarified SNPE-A behavior, and a couple of bug fixes.
1
0
3
🎉 Just in time for the end of the year, we’ve released a new version of sbi!. 📦 v0.23.3 comes with exciting features, bug fixes, and docs updates to make SBI smoother and more robust. Check it out! 👇. 🔗 Full changelog:
github.com
v0.23.3 Highlights 🤩 docs: Add conda-forge install instructions by @matthewfeickert in #1340 feat: NLE with multiple iid conditions by @janfb in #1331 What's Changed 🚧 fix: Correted typo in ...
1
0
5
We are launching an SBI Discord Server! 🎉 . We want to use this server to further build a community around SBI, i.e., for sharing insights, questions and events around SBI in general and the sbi package in particular. You are all invited to join! 🤗
github.com
Dear all, we are launching an SBI Discord Server! 🎉 We want to use this server to further build a community around SBI, i.e., for sharing insights, questions and events around SBI in general and...
0
0
4
As of today, 61 scientists and engineers have contributed to the sbi toolbox. We are extremely happy about this, and want to give a huge **thank you** to all contributors! Stay tuned for more info on events and updates!
github.com
sbi is a Python package for simulation-based inference, designed to meet the needs of both researchers and practitioners. Whether you need fine-grained control or an easy-to-use interface, sbi has ...
0
0
3
For all steps of the inference process, the sbi toolbox supports strong defaults if needed, but also provides full flexibility if desired. For example, you can use a pre-configured training loop, or you can write it yourself.
1
0
2
The sbi toolbox implements a wide range of simulation-based inference methods. It implements NPE, NLE, and NRE (all amortized or sequential), modern neural networks (flows, flow-matching, diffusion models), samplers, and diagnostic tools.
1
0
1
The sbi package is growing into a community project 🌎 To reflect this and the algorithms, neural nets, and diagnostics that have been added since its initial release, we have written a new software paper. Reach out if you want to get involved:
arxiv.org
Scientists and engineers use simulators to model empirically observed phenomena. However, tuning the parameters of a simulator to ensure its outputs match observed data presents a significant...
1
11
32
Finally, a huge thank you to all contributors who helped make this possible and who joined us for the hackathon in Tübingen! Stay tuned for more events like this in 2025 🥁 8/8.
0
0
8
The new `sbi` version is already available on PyPI: `pip install sbi --upgrade`. We would love to hear what is (or is not) working for you. Many of the above features were based on previous user feedback, so please let us know what you think! Happy coding! 7/8.
1
0
5
There are so many more cool new things: Support for zuko flows, an interface to PyMC, ABC methods for trial-based data, and a new plotting function. We also reworked the documentation website to make it easier to navigate! See here for all changes: 6/8.
github.com
sbi is a Python package for simulation-based inference, designed to meet the needs of both researchers and practitioners. Whether you need fine-grained control or an easy-to-use interface, sbi has ...
1
1
6
`sbi` now also supports batched sampling for NPE and for MCMC (for NLE, NRE). This allows to largely speed up sampling for many observations when running amortized inference. 5/8
1
0
5
You are worried whether your posteriors are correct? The new `sbi` version has you covered! In addition to the already existing SBC and coverage checks, `sbi` now also supports L-C2ST and TARP. Tutorial: 4/8
1
0
5
Score-matching and flow-matching have been demonstrated to perform well on a series of inference problems. The sbi toolbox now supports both of them! Tutorial: 3/8
1
0
7
The new sbi version allows to fully expose the training loop for expert users, giving complete control over every step of the training and inference process. Tutorial: 2/8
1
1
7