Ab-initio job: Detected Nan values in engine.error 31337280 in total. 90 particles with Nan

3D Reconstruction
1

I got that error launching the ab-initio after many iterations. What could it be due to?

2

Please can you post

  1. the version of CryoSPARC you are using
  2. the final 30 lines of the event log
  3. the non-default parameters of the failed job
  4. the results of a Check for Corrupt Particles job with the Check for NaN values option
3

1
11660Ă—428 63.2 KB

Hi, I am encountered with the same problem, and the particle stack is from EMPIAR-10278. All the parameters of Ab-initio job are default. The Check for Corrupt Particles job with the Check for NaN values option ran successfully. And can you explain that why my Ab-initio job fail? Thanks.(the version:v4.6.0)

1 Like
4

Hi @chengyr1999,

The check for corrupt particles job doesn’t fail if it founds NaN values, it just emits a warning, and a short summary at the end of the stream log which lists any files that had NaN values. If you’re able to, could you please check the log of the check for corrupt particles job for any such warnings?

– Harris

5

Hi @chengyr1999,

Your input would be very valuable to us as we try to identify the root cause of these NaN value issues. If you have time to respond to my previous query, we would be grateful.

–Harris

6

hello Harris @hsnyder, I guess I have the same issue - The 2D runs fine but any 3D give the following error - [2025-05-29 17:15:22.27]
[CPU: 1.25 GB Avail: 501.96 GB]
Traceback (most recent call last):
File “/home/cryosparc_user/cryosparc/cryosparc_worker/cryosparc_compute/jobs/runcommon.py”, line 2306, in run_with_except_hook
run_old(*args, **kw)
File “cryosparc_master/cryosparc_compute/gpu/gpucore.py”, line 136, in cryosparc_master.cryosparc_compute.gpu.gpucore.GPUThread.run
File “cryosparc_master/cryosparc_compute/gpu/gpucore.py”, line 137, in cryosparc_master.cryosparc_compute.gpu.gpucore.GPUThread.run
File “cryosparc_master/cryosparc_compute/engine/engine.py”, line 1146, in cryosparc_master.cryosparc_compute.engine.engine.process.work
File “cryosparc_master/cryosparc_compute/engine/engine.py”, line 431, in cryosparc_master.cryosparc_compute.engine.engine.EngineThread.compute_error
ValueError: Detected NaN values in engine.compute_error. 7933671 NaNs in total, 215 particles with NaNs. Any ideas? Thank you. Regards, Dmitry P.s. I run the check particles protocol with no issues

7

@Dmitry What was the CryoSPARC version for this job?

8

Hello, @wtempel v4.7. Regards, Dmitry

9

@Dmitry

Please can you let us know

  1. a list the 3D job types where you observed the error
  2. Do jobs of these types always fail?
  3. Are the tracebacks/referenced code files and line numbers the same for all these job types?
  4. Do repeated reruns of an affected ab initio fail consistently/always?
1 Like
10

Hello @wtempel, I can look more closer, but what eventually helped is to change the reference in the refinement. Kind regards,
Dmitry