search

scipion-em / scipion-em-aretomo

Plugin to use AreTomo within the Scipion framework
GNU General Public License v3.0
2 stars 0 forks source link

Jj output ctf #62

Closed JorMaister closed 9 months ago

azazellochg commented 9 months ago

Thank you very much, @JorMaister that's a lot of work! I will test this when I can

JorMaister commented 9 months ago

I'm changing the state to draft until I add the psd file and make the tests check for it and also a CUDA_LIB variable (I forgot that couple of things). I expect them to be tomorrow in the morning.

azazellochg commented 9 months ago

No worries, I'll try to review it this weekend.

azazellochg commented 9 months ago

I already have problems with the first test: align&rec with ctf, 3 threads, 2 TS, 1 GPU. The first aretomo process launches and somewhere in the middle of the run the second TS is started on the same GPU. This slows down processing tremendously. I cannot reduce the number of threads <3.

azazellochg commented 9 months ago

ok. If you can make a shorter comment :) CTF est requires Cs, the rest is required for dose-weighting Best regards, Grigory

Grigory Sharov, Ph.D.

MRC Laboratory of Molecular Biology, Francis Crick Avenue, Cambridge Biomedical Campus, Cambridge CB2 0QH, UK. tel. +44 (0) 1223 267228 <+44%201223%20267228> e-mail: @.***

On Mon, Feb 12, 2024 at 9:08 AM Jorge Jiménez de la Morena < @.***> wrote:

@.**** commented on this pull request.

@azazellochg https://github.com/azazellochg. why did you remove the comment regarding the CTF estimation? I think it's not clear why the behaviour relies on the Cs, so it should be explained somehow

— Reply to this email directly, view it on GitHub https://github.com/scipion-em/scipion-em-aretomo/pull/62#pullrequestreview-1874779090, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABVBPJXHPNWL32IFWFXHHATYTHLXHAVCNFSM6AAAAABC3PQHJCVHI2DSMVQWIX3LMV43YUDVNRWFEZLROVSXG5CSMV3GSZLXHMYTQNZUG43TSMBZGA . You are receiving this because you were mentioned.Message ID: @.***>

JorMaister commented 9 months ago

I already have problems with the first test: align&rec with ctf, 3 threads, 2 TS, 1 GPU. The first aretomo process launches and somewhere in the middle of the run the second TS is started on the same GPU. This slows down processing tremendously. I cannot reduce the number of threads <3.

Hi, @azazellochg. Pablo is currently working on it.

JorMaister commented 9 months ago

somewhere in the middle of the run the second TS is started on the same GPU. This slows down processing tremendously. I cannot reduce the number of threads <3.

Hi, @azazellochg. After the last changes in pyworkflow, it's now solved. I'll change the default number of threads to 2, as the default value for the GPU is one GPU, thus there will be 1 thread for the step generator and the other for running the steps.