Git Product home page Git Product logo

Comments (5)

mhhennig avatar mhhennig commented on June 13, 2024

My guess is that the bottleneck is likely the function collecting spike cut-outs (both used in curation and phy export). What format is your raw data file? Also, do you apply any filters on the recording extractor?

from spikeinterface.

guitchounts avatar guitchounts commented on June 13, 2024

The raw data was acquired with Open Ephys GUI, stored in their binary format. I'm applying bandpass filtering and common mode rejection on the recording extractor, and using the cmr'd extractor for sorting, curation, phy, etc.

Is it possible to speed up the cut-out step somehow? or at least to save the cutouts in one step so they don't have to be re-computed several times?

from spikeinterface.

samuelgarcia avatar samuelgarcia commented on June 13, 2024

in spike interface there is a cache system to avoid to recompute the preprocessing chain and extractor several times waveforms snippet (very long) @alejoe91 do we have a notebook example for the cache workflow ?

from spikeinterface.

alejoe91 avatar alejoe91 commented on June 13, 2024

Hi @guitchounts

there are a few things you can do:

  1. Cache the recording_cmr (VERY IMPORTANT): this dumps a binary version of your recording (in your case bandpass filtered + cmr) to a dat binary file. You can simply do: recording_cmr = se.CacheRecordingExtractor(recording_cmr, 'recording_cmr.dat'). The process will take a few minutes, but it'll speed up everything else (otherwise each single waveforms is filtered separately later!)

  2. Use multiple jobs: when computing the SNR (and therefore the waveforms in the background) you can use the n_jobs=8 (or 4-16...). This will compute waveforms and templates in parallel. Same holds for the export_to_phy.

  3. Note that the export to phy is slower because ALL waveforms need to be extracted to compute amplitudes and PC scores. If you are not interested in those (you'll still be able to look at templates, similarity, and cross-correlograms), you can try the to set the compute_amplitudes=False and compute_pc_features=False when calling the export_to_phy

Hope this helps! :)

from spikeinterface.

alejoe91 avatar alejoe91 commented on June 13, 2024

Potprocessing efficiency issues are solved with latest releases of spiketoolkit

from spikeinterface.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.