Package for plotting GPRegression results

pygp plotting tools

Tools to plot Gaussian process pygp.gp regression output.

pygp.plot.gpr_plot.plot_sausage(X, mean, std, alpha=None, format_fill={'alpha': 0.1, 'facecolor': 'k'}, format_line={'color': 'g', 'lw': 3, 'ls': 'dashed', 'alpha': 1}, ax=None)

plot saussage plot of GP. I.e:

_images/sausage.png
returns: : [fill_plot, line_plot]
The fill and the line of the sausage plot. (i.e. green line and gray fill of the example above)

Parameters:

X : [double]
Interval X for which the saussage shall be plottet.
mean : [double]
The mean of to be plottet.
std : [double]
Pointwise standard deviation.
format_fill : {format}
The format of the fill. See http://matplotlib.sourceforge.net/ for details.
format_line : {format}
The format of the mean line. See http://matplotlib.sourceforge.net/ for details.
pygp.plot.gpr_plot.plot_training_data(x, y, shift=None, replicate_indices=None, format_data={'marker': '.', 'alpha': 0.5, 'markersize': 9, 'linestyle': '--', 'lw': 1}, draw_arrows=0, plot_old=False)

Plot training data input x and output y into the active figure (See http://matplotlib.sourceforge.net/ for details of figure).

Instance plot without replicate groups:

_images/plotTraining.png

Instance plot with two replicate groups and a shift in x-koords:

_images/plotTrainingShiftX.png

Parameters:

x : [double]
Input x (e.g. time).
y : [double]
Output y (e.g. expression).
shift : [double]
The shift of each replicate group.
replicate_indices : [int]
Indices of replicates for each x, rexpectively
format_data : {format}
Format of the data points. See http://matplotlib.sourceforge.net/ for details.
draw_arrows : int
draw given number of arrows (if greator than len(replicate) draw all arrows. Arrows will show the time shift for time points, respectively.

Table Of Contents

Previous topic

Package for Priors of Gaussian Processes

Next topic

Application Example of GP regression

This Page