Skip to content

Support per-iteration CTF correction - #138

Open
amineuron wants to merge 1 commit into
warpem:mainfrom
amineuron:feat/adaptive-ctf-clean
Open

Support per-iteration CTF correction#138
amineuron wants to merge 1 commit into
warpem:mainfrom
amineuron:feat/adaptive-ctf-clean

Conversation

@amineuron

@amineuron amineuron commented Jul 25, 2026

Copy link
Copy Markdown
Contributor

Allows apply_ctf to be overridden in each iteration_settings entry. The resolved value is used consistently for training reconstructions and alignment, and both phases report it in the log.

CLIMP validation (7 tilt series): adaptive CTF on the final two iterations ran in 8:56:37 vs 9:38:26 with CTF throughout (7.2% faster). Local-warp RMS / trajectory roughness were 25.62 / 19.60 Å adaptive, 26.39 / 21.04 Å CTF-off, and 21.44 / 17.98 Å CTF-on.

Tests: 173 passed, 2 skipped.

@McHaillet

Copy link
Copy Markdown
Collaborator

Hi Amin, in principal I'd be happy to merge this in, but you need to give me some more explanation of the tests that you ran. When we tried this we never noticed a big difference with CTF turned on or off. What is the test that you ran? How did you validate? Did you compare against some ground truth or STA?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants