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 you covered.

749 stars 201 forks 749 watchers Python Apache License 2.0
bayesian-inference likelihood-free-inference machine-learning parameter-estimation pytorch simulation-based-inference
3 Open Issues Need Help Last updated: Nov 5, 2025

Open Issues Need Help

View All on GitHub

AI Summary: This GitHub issue reports a `TypeError` occurring when using `pymc` for sampling since `pymc` version `5.20.1`. The proposed solution is to monitor for a new `pymc` release that addresses the problem and then remove the `fail` marks from specific tests in the `sbi` codebase once the issue is resolved upstream.

Complexity: 1/5
bug good first issue hackathon

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 you covered.

Python
#bayesian-inference#likelihood-free-inference#machine-learning#parameter-estimation#pytorch#simulation-based-inference
GPU compatible (L-)C2ST about 3 hours ago
enhancement good first issue

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 you covered.

Python
#bayesian-inference#likelihood-free-inference#machine-learning#parameter-estimation#pytorch#simulation-based-inference

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 you covered.

Python
#bayesian-inference#likelihood-free-inference#machine-learning#parameter-estimation#pytorch#simulation-based-inference