In machine learning, what is definition of “downstream”

machine learningscikit-learn

I've read the [PCA documentation](http://scikit-learn.org/stable/modules/generated/sklearn.decomposition.PCA.html
) of scikit-learn.

[…] improve the predictive accuracy of the downstream estimators […]

What is the definition of “downstream” in machine learning?

Best Answer

I know the term "downstream" from neural networks. In those machine learning algorithms, you have so called "neurons" which are usually in form of a DAG. Downstream is everything after a certain neuron. You say neuron y is downstream of neuron x if and only if there is a directed path from x to y.

In a more general setting, I can only guess: y is downstream of x if and only if y uses data processed by x.

Related Solutions

Python – Factor Loadings using sklearn

Multiply each component by the square root of its corresponding eigenvalue:

pca.components_.T * np.sqrt(pca.explained_variance_)

This should produce your loading matrix.

Python – In python, make a tempfile in the same directory as another file

You can use:

prefix to make the temporary file begin with the same name as the original file.
dir to specify where to place the temporary file.
os.path.split to split the directory from the filename.

import tempfile
import os
dirname, basename = os.path.split(filename)
temp = tempfile.NamedTemporaryFile(prefix=basename, dir=dirname)
print(temp.name)

Best Answer

Related Solutions

Python – Factor Loadings using sklearn

Python – In python, make a tempfile in the same directory as another file

Related Topic