FastICA | python version of fast and robust ICA | Build Tool library
kandi X-RAY | FastICA Summary
kandi X-RAY | FastICA Summary
A python version of fast and robust ICA based on the paper of Aapo Hyvärinen.
Support
Quality
Security
License
Reuse
Top functions reviewed by kandi - BETA
Currently covering the most popular Java, JavaScript and Python libraries. See a Sample of FastICA
FastICA Key Features
FastICA Examples and Code Snippets
Community Discussions
Trending Discussions on FastICA
QUESTION
I'm following the answer to this question and this scikit-learn tutorial to remove artifacts from an EEG signal. They seem simple enough, and I'm surely missing something obvious here.
The components extracted don't have the same length as my signal. I have 88 channels of several hours of recordings, so the shape of my signal matrix is (88, 8088516). Yet the output of ICA is (88, 88). In addition to being so short, each component seems to capture very large, noisy-looking deflections (so out of 88 components only a couple actually look like signal, the rest look like noise). I also would have expected only a few components to look noisy. I suspect I'm doing something wrong here?
The matrix of (channels x samples) has shape (88, 8088516).
Sample code (just using a random matrix for minimum working purposes):
...ANSWER
Answered 2022-Mar-24 at 10:13You need to run the fit_transform
on the transpose of your samples_matrix
instead of the samples_matrix
itself (so provide a 8088516 x 88 matrix instead of an 88x8088516 to the method).
QUESTION
I have created a function that run FastICA
on a dataset with different number of components and it returns ICA signals (S matrix) but in a long format.
ANSWER
Answered 2022-Feb-06 at 08:50I cannot reproduce the error, however, I think you misunderstood what
mapply
does.
When you apply mapply
to a function FUN
, with arguments that are
lists or vectors (consider that R basic types like numbers or characters are always vectors), the function FUN
is called iteratively on the first element of all the arguments, then on the second, etc. Arguments are recycled if necessary.
For instance:
QUESTION
I updated Neuraxle to the latest version (3.4).
I noticed the whole auto_ml.py
was redone. I checked the documentation but there is nothing about it. On git it seems method RandomSearch()
was replaced a long time ago by AutoML()
method. However the parameters are different.
Does somebody knows how can I channel Boston Housing example pipeline to automatic parameter search in latest Neuraxle version (3.4)?
...ANSWER
Answered 2020-May-16 at 02:18Here is a solution to your problem, this is a new example that isn't yet published on the documentation site:
- https://drive.google.com/drive/u/0/folders/12uzcNKU7n0EUyFzgitSt1wSaSvV4qJbs (go see the solution to the 2nd coding Kata from there)
Sample pipeline code from the link above:
QUESTION
I'm using the FastIca toolbox (https://research.ics.aalto.fi/ica/fastica/) but am confused about the orientation of the resulting W (separating/unmixing) matrix.
Let X be a n x B matrix where n is the number of signals in a data set and B is the number of time points sampled at.
I've been calculating the W matrix using:
...ANSWER
Answered 2020-Apr-28 at 17:53It should be Y=W*X
. To be sure, you can reduce the number of component to estimate and then W
should no longer be square:
QUESTION
I have a code example - sklearn pipeline that has two components (PCA and Random Forest), I want to use the intermediate results of the pipeline in order to bring some explainability. I know that it is possible to use .get_params() to see the intermediate steps, but is it possible to save or extract the intermediate results for additional actions? I want to apply additional functions of the PCA (1.1. and 1.2 sections in the code)
...ANSWER
Answered 2020-Feb-06 at 19:51We can assign get_params()
to a variable which should return an object of type sklearn.decomposition.pca.PCA
. With this, we are able to access all the methods and attributes of the decomposition.
Community Discussions, Code Snippets contain sources that include Stack Exchange Network
Vulnerabilities
No vulnerabilities reported
Install FastICA
You can use FastICA like any standard Python library. You will need to make sure that you have a development environment consisting of a Python distribution including header files, a compiler, pip, and git installed. Make sure that your pip, setuptools, and wheel are up to date. When using pip it is generally recommended to install packages in a virtual environment to avoid changes to the system.
Support
Reuse Trending Solutions
Find, review, and download reusable Libraries, Code Snippets, Cloud APIs from over 650 million Knowledge Items
Find more librariesStay Updated
Subscribe to our newsletter for trending solutions and developer bootcamps
Share this Page