Redacted excerpt; full evidence withheld.

$ cd <guest-path> && pwd && ls -la && git status 2>&1 | head -3; python3 --version; cat README.md 2>/dev/null | head -c 7000
exit: 0
em one way and you have flow matching with an independent coupling. Another way and you have OT-coupled rectified flow. Another and you have posterior sampling for an inverse problem, reusing the very same pretrained field.

Full documentation lives on the [**DeltaFlow website**](https://phrugsa-limbunlom.github.io/deltaflow).

## What's inside

### Probability paths (`deltaflow.interpolants`)

The path decides how noise becomes data. Every path returns the interpolated state $x_t$ and its conditional target velocity $u_t$.

[excerpt truncated]