Konjugate v0.7.0: Causal Inference

Causal inference

Import a CSV of multivariate time-series data and Konjugate proposes the causal graph behind it: which columns drive which, with what lag, and whether the relationship is linear or curved.

• Two-stage pipeline: a lagged partial-correlation skeleton pass finds which variables are related at all, then a joint ridge regression fit determines the direction of each relationship and fits it as a linear or polynomial equation.
• Three nonlinearity modes — "Linear only," "Allow curvature up to degree N," or "Let the tool decide" (compares linear vs. polynomial per relationship and keeps whichever fits better).
• Every proposed node and edge is reviewable — checked in or out — before it's added to your model, through the same pipeline the AI assistant uses.
• Imported nodes default to a sphere shape and auto-arrange in a circle sized to their count, so edges stay legible instead of overlapping.

Fixed

• Pose Visualizer's playback could report itself finished a frame before the timeline had actually caught up, due to a race between a seek request and its confirmation.

Installation notes: Konjugate isn't code-signed yet (an active choice while the project is in alpha), so your OS may show a security warning on first launch. This doesn't mean the download is corrupted.

• macOS — if you see "is damaged and can't be opened": move Konjugate.app to Applications, then in Terminal run xattr -cr /Applications/Konjugate.app. Or: System Settings → Privacy & Security → Open Anyway.
• Windows — if you see a blue "Windows protected your PC" warning from SmartScreen, click More info, then click the Run anyway button that appears.

Full Changelog: v0.6.6...v0.7.0