Welcome to pydiodon’s documentation!

A python library for linear dimension reduction.

Identity card

authors
  • Alain Franc

  • Jean-Marc Frigerio

mail
contributors
  • Olivier Coulaud

  • Romain Peressoni

  • Florent Pruvost

maintainer

Alain Franc

started

21/02/17

version

23.06.12

release

0.0.4

licence

GPL-3.0 or later

This file is part of diodon project. You can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

Information: the development of this library still is ongoing. When a function has not been tested or its development still in infancy, this is indicated in its documentation, and it is advised not to use it.

Functions

coa

Correspondance Analysis of an array

mds

Multidimensional Scaling of a distance or dissimilarity array

pca

Principal Component Analysis

mat_evd

Computes the eigenvalues and eigenvectors of a matrix

svd_grp

SVD of a (large) matrix by Gaussian Random Projection

bicentering

Bicentering a matrix

centering_operator

Centering operator for a matrix

center_col

centers a matrix columnwise

center_row

centers a matrix row-wise

scale

Scales a matrix columnwise

get_correlation_matrix

Gets the correlation matrix

dis2gram

Computes a Gram matrix knowing a distance matrix

pca_core

core method for PCA (Principal Component Analysis) of an array

pca_met

PCA of an array with metrics on rows and/or columns - without guaranty

pca_iv

PCA of an array with instrumental variables

project_on

Builds the projection operator on space spanned by the columns of an array

loadfile

A generic function for loading datasets as numpy arrays

load_ascii

Loads an ascii file as a numpy array

load_hdf5

loads a hdf5 file as a numpy array

load_dataset

Loads datasets provided as examples after having cloned companion git of (py)diodon.

writefile

A generic function for saving a numpy array as a file

write_ascii

Write an ascii file

write_hdf5

Write a hdf5 file

plot_eig

Plots the singular or eigenvalues

plot_components_scatter

Scatter plot of the result of a Principal Component Analysis

plot_components_splines

Plots principal components as lines smoothed with cublic splines, one per item

plot_cumulated_quality_per_item

plots the quality per item

plot_components_quality

Scatter plot of the principal components with each dot colored according to the cumulated quality of the item for a given axis

plot_var

Plots the correlations between the variables

plot_var_heatmap

Plots a heatmap of the coordinates of the principal axis

plot_correlation_matrix

Plots a heatmap of the correlation matrix

plot_coa

Scatter plot of the result of a Correspondence Analysis

hovering

Display a label of one dot selected with the mouse in a scatter plot

plot_components_heatmap

Builds the density heatmap of the point cloud of the components of a PCA

Indices and tables