search

IsoNet-cryoET / IsoNet

Self-supervised learning for isotropic cryoET reconstruction
https://www.nature.com/articles/s41467-022-33957-8
MIT License
70 stars 13 forks source link

Isonet stops after a few iterations at time for refinement #47

Open jake4844 opened 1 year ago

jake4844 commented 1 year ago

Hi,

I have been having recent issues with isonet where it will run fine for a few iterations during refinement and at some during Noise level estimation it will just stop and quit with no errors. Is there something causing this to happen? Thanks!

procyontao commented 1 year ago

Could you tell whether it stopped only at 10 iter when noise volume are generated, or this could happen at any iteration. Could you share your log file?

jake4844 commented 1 year ago

log.txt Here is the log file! Let me know what you find out:)

XiaJia98 commented 1 year ago

Could you tell whether it stopped only at 10 iter when noise volume are generated, or this could happen at any iteration. Could you share your log file?

hi, I have run in this problem, the refine step stopped only at 10 itr when noise volume are generated and reported error "TypeError: iradon() got an unexpected keyword argument 'filter_name'". Is there any solution? Thanks so many!

procyontao commented 1 year ago

Hi,

Sorry for late reply. I think you are using a old version of scikit-image package. "filter_name" parameter was introduced after skimage 0.17.x

XiaJia98 commented 1 year ago

Hi,

Sorry for late reply. I think you are using a old version of scikit-image package. "filter_name" parameter was introduced after skimage 0.17.x

thanks，it has worked well now!

KASochacki commented 10 months ago

Hi, I believe I am having a similar problem where the refine step stops after iteration 10. The difference is that my error is local variable 'num_noise_volume' referenced before assignment.

skimage.version '0.17.2' numpy.version '1.19.5' Is there any chance you are familiar with this problem? Thank you!

KASochacki commented 10 months ago

This ended up just being that an initial memory problem resulted in an empty training_noise directory. Further attempts to run the refinement ended with the num_noise_volume error. When the training_noise directory was removed, the error went away and the insufficient memory problem became more clear.

richyjunior commented 1 month ago

This ended up just being that an initial memory problem resulted in an empty training_noise directory. Further attempts to run the refinement ended with the num_noise_volume error. When the training_noise directory was removed, the error went away and the insufficient memory problem became more clear.

Hi, Do yo solve this problem? I’m facing a similar issue when running my code. The program stops with a num_noise_volume error when I try to run it using the continue_from command."