Replies: 1 comment
-
|
Hi @mkmellon5 , that warning checks to see if your training particle coordinates fill most of the area of your images. The most common mistake to raise this would be not scaling the coordinates to match your images if you have downsampled/binned i.e. images that are 5000x5000 but coordinates for 500x500. If you’re sure that the scale is correct (you should visualize the images with coordinates overlaid to confirm), then you can safely ignore it. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I am trying to utilize Topaz Train, and I've been working through one error after another.
My particles are from the selected 2D classes and the micrographs are from Extract from Micrographs. I have the correct path to Topaz executable (I got a different error when that was wrong). And the absolute path of directory containing preprocessed directory is to the patch motion correction job, which is the only folder that prevents a different error from occurring. During patch. Motion correction I did have the output f-crop factor set to 1/2.
The following are the parameters I changed from default for topaz train: estimated particle diameter, expected # of particles, number of epochs, and turned off ‘use denoised micrographs’.
The warning I am getting during the run is:
"# WARNING: no coordinates are observed with x_coord > 498 or y_coord > 390 . Did you scale the micrographs and particle coordinates correctly?
[2025-07-08 13:30:28.23] [CPU: 255.0 MB Avail: 86.15 GB]"
What am I missing? Any guidance is appreciated. None of chat GPTs suggestions are helping me get past this hurdle.
Beta Was this translation helpful? Give feedback.
All reactions