Julia Linhart
Julia Linhart
## What does this implement/fix? Explain your changes L-C2ST(-NF) diagnostic: class, tutorial and tests. ## Does this close any currently open issues? Fixes #1005 ## Any relevant code examples, logs,...
** Problem Description ** Similar to issue #1085 fixed in PR #1086 for the inverse-transform method. It can be useful to access the base distribution of the normalizing flow or...
**General Goal:** Implementation for the `l-c2st` method from [L-C2ST: Local Diagnostics for Posterior Approximations in Simulation-Based Inference](https://arxiv.org/abs/2306.03580). The existing metrics (`c2st`, `mmd`, etc.) are implemented in the `utils/metrics.py` file. They...
## **Is your feature request related to a problem? Please describe.** When working with high-dimensional datasets, computing C2ST (and particularly L-C2ST) metrics can be quite slow. This is primarily due...
There is no need to run `get_data` if the results have already been computed and will be loaded from the cache. We want to avoid calling `get_data` if not required,...
**Summary:** A `zsh: segmentation fault` error is raised when importing `pandas` after pip-installing `benchopt` on a macOs device with a M1 (ARM) processor. **Explanation:** This error is due to the...