pwstegman / bci.js Goto Github PK
View Code? Open in Web Editor NEW:bar_chart: EEG signal processing and machine learning in JavaScript
Home Page: https://bci.js.org
License: MIT License
:bar_chart: EEG signal processing and machine learning in JavaScript
Home Page: https://bci.js.org
License: MIT License
Welch's method for power spectral density estimation.
Implemented and set for release in v1.8. Need to add unit tests.
I am loading an edf using:
bci.loadEDF
In return I am getting an object that contains a channel property. Each channel contains samples, I am passing each of these samples to bci.averageBandPowers but getting this error:
TypeError: Cannot read property 'length' of undefined
at signalBandPower (C:\Users\HP\Desktop\projects\driverDrowsiness\node_modules\bcijs\lib\math\signalBandPower.js:49:41)
at Object.averageBandPowers (C:\Users\HP\Desktop\projects\driverDrowsiness\node_modules\bcijs\lib\math\averageBandPowers.js:25:29)
at getData (C:\Users\HP\Desktop\projects\driverDrowsiness\index.js:11:37)
at process._tickCallback (internal/process/next_tick.js:68:7)
at Function.Module.runMain (internal/modules/cjs/loader.js:834:11)
at startup (internal/bootstrap/node.js:283:19)
at bootstrapNodeJSCore (internal/bootstrap/node.js:622:3)
P.S In channels.sample I am getting a 1d array rather than the 2d one in bci.loadedf
This is my edf file
Multitaper method for power spectral density estimation.
Implemented in v1.8. Need to add unit tests.
Edit: Moved to v1.9
For computation of discrete prolate spheroidal (Slepian) sequences, a length up to 128 is calculated. Lengths past 128 are linearly interpolated up.
For tapers with near-unity eigenvalues for common time half bandwidths (2.5, 4, etc.) the curves are smooth enough that linear interpolation works well. A more efficient eigenvector method will help remove the need to interpolate.
When using a small FFT size, the PSD returns NaN
.
Steps to reproduce.
bci.signalBandPower([1, 2, 3, 4, 5, 6, 7], 8, 256, 'alpha')
=> NaN
bci.signalBandPower([1, 2, 3, 4, 5, 6, 7], 16, 256, 'alpha')
=> NaN
bci.signalBandPower([1, 2, 3, 4, 5, 6, 7], 32, 256, 'alpha')
=> 26.416893639442556
When attempting to pack this code for use on a browser-only solution, webbci's dependence on node-osc causes webpack to error due to node-only dependencies.
ERROR in ./node_modules/node-osc/lib/safeDgram.js
Module not found: Error: Can't resolve 'dgram' in './node_modules/node-osc/lib'
@ ./node_modules/node-osc/lib/safeDgram.js 3:12-28
@ ./node_modules/node-osc/lib/Server.js
@ ./node_modules/node-osc/lib/index.js
@ ./node_modules/webbci/lib/network.js
@ ./node_modules/webbci/index.js
If you don't expect for this code to run on the browser side only, then this is probably expected behavior.
This computes average power across channels as expected:
let power = bci.bandpower(samples, samplerate, ['alpha'], {average: true});
But this does not average across channels ({average: true}
is ignored)
let power = bci.bandpower(samples, samplerate, 'alpha', {average: true});
Hello Pierce!
My name is Anton and I’m using your js package called BCI.js but I can’t understand one thing and I believe you can help me :)
I have an array of signals from EEG session which lasted for 30 seconds and I want to get average alpha band power for every second.
I divide my signals by 30 and on each subarray I’m using signalBandPower method. Is this correct or I’m doing something wrong?
Thank you in advance!
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.