Note

Go to the end to download the full example code.

SKADA logo generation

In this example we plot the logos of the SKADA toolbox.

This logo is that it is done 100% in Python and generated using matplotlib and plotting the solution of the EMD solver from POT.

# Author: Remi Flamary
#
# License: BSD 3-Clause
# sphinx_gallery_thumbnail_number = 1

import matplotlib.pyplot as plt
import numpy as np
import pylab as pl

plt.rcParams.update({"text.usetex": True, "font.family": "Computer Modern"})


def get_lda_interp(xs, xt, alpha, len_=1):
    """Compute the LDA interpolation between two domains"""
    # compute means (classes assumed to be balanced)
    m1 = (1 - alpha) * xs[: n // 2, :].mean(0) + (alpha) * xt[: n // 2, :].mean(0)
    m2 = (1 - alpha) * xs[n // 2 :, :].mean(0) + (alpha) * xt[n // 2 :, :].mean(0)

    vo = m2 - m1
    vo = vo / np.linalg.norm(vo)

    theta = np.pi / 2
    rot = np.array([[np.cos(theta), -np.sin(theta)], [np.sin(theta), np.cos(theta)]])

    vor = rot.dot(vo)

    mm = 0.5 * m1 + 0.5 * m2

    return np.vstack((mm - len_ * vor, mm + len_ * vor))

n = 40

np.random.seed(42)

xs = 0.2 * np.random.randn(n, 2)

# class  specific change
xs[n // 2 :, 0] += 0.5
xs[n // 2 :, 1] -= 1

# classes 0 an 3 for blue/red colors
ys = np.zeros(n)
ys[n // 2 :] = 3

# global changes
xt = 0.15 * np.random.randn(n, 2)
xt[:, 0] += 0.7
xt[:, 1] -= 0

# class specific change
xt[n // 2 :, 0] += 1
xt[n // 2 :, 1] -= 0.2

# class 7 for gray color (target without label)
yt = np.ones(n) * 7

Plot the full logo in white

# target data for full logo
xt2 = xt.copy()
xt2[:, 0] += 6.3


nb = 10
alpha0 = 0.2
alphalist = np.linspace(0, 1, nb)

pl.figure(2, (9, 1.5))
pl.clf()
alpha = 0.7

# plot samples
pl.scatter(
    xs[ys == 0, 0],
    xs[ys == 0, 1],
    c="w",
    cmap="tab10",
    vmax=10,
    alpha=alpha,
    label="Source data",
    marker="o",
)
pl.scatter(
    xs[ys == 3, 0],
    xs[ys == 3, 1],
    c="w",
    cmap="tab10",
    vmax=10,
    alpha=alpha,
    label="Source data",
    marker="s",
)
pl.scatter(
    xt2[ys == 0, 0],
    xt2[ys == 0, 1],
    c="w",
    cmap="tab10",
    vmax=10,
    alpha=alpha,
    label="Source data",
    marker="o",
)
pl.scatter(
    xt2[ys == 3, 0],
    xt2[ys == 3, 1],
    c="w",
    cmap="tab10",
    vmax=10,
    alpha=alpha,
    label="Source data",
    marker="s",
)

# plot classifiers
ax = pl.axis()
for i in range(nb):
    xclass = get_lda_interp(xs, xt2, alphalist[i], 2)
    pl.plot(
        xclass[:, 0],
        xclass[:, 1],
        color="w",
        alpha=alpha0 + (1 - alpha0) * alphalist[i],
        zorder=0,
        linewidth=2,
    )
pl.text(
    1.3,
    -1.18,
    r"\bf\textsf{SKADA}",
    fontsize=80,
    usetex=True,
    zorder=0.5,
    color="w",
    alpha=0.9,
)

pl.axis(ax)
pl.axis("off")

pl.savefig("skada_logo_full_white.png", transparent=True, bbox_inches="tight", dpi=400)
pl.savefig("skada_logo_full_white.svg", transparent=True, bbox_inches="tight", dpi=400)
plot skada logo
/home/circleci/project/examples/plot_skada_logo.py:256: UserWarning: No data for colormapping provided via 'c'. Parameters 'cmap', 'vmax' will be ignored
  pl.scatter(
/home/circleci/project/examples/plot_skada_logo.py:266: UserWarning: No data for colormapping provided via 'c'. Parameters 'cmap', 'vmax' will be ignored
  pl.scatter(
/home/circleci/project/examples/plot_skada_logo.py:276: UserWarning: No data for colormapping provided via 'c'. Parameters 'cmap', 'vmax' will be ignored
  pl.scatter(
/home/circleci/project/examples/plot_skada_logo.py:286: UserWarning: No data for colormapping provided via 'c'. Parameters 'cmap', 'vmax' will be ignored
  pl.scatter(

Total running time of the script: (0 minutes 0.948 seconds)

Gallery generated by Sphinx-Gallery