def test_pca_dim():
"""
"""
n, p = 100, 5
X = randn(n, p)*.1
X[:10] += np.array([3, 4, 5, 1, 2])
pca = PCA(n_comp='mle')
pca.fit(X)
assert_true(pca.n_comp == 1)
def test_infer_dim_2():
"""
"""
n, p = 1000, 5
X = randn(n, p)*.1
X[:10] += np.array([3, 4, 5, 1, 2])
X[10:20] += np.array([6, 0, 7, 2, -1])
pca = PCA(n_comp=p)
pca.fit(X)
spect = pca.explained_variance_
assert_true(_infer_dimension_(spect, n, p) > 1)
def test_pca_check_projection():
"""test that the projection of data is correct
"""
n, p = 100, 3
X = randn(n, p) * .1
X[:10] += np.array([3, 4, 5])
pca = PCA(n_comp=2)
pca.fit(X)
Xt = 0.1* randn(1, p) + np.array([3, 4, 5])
Yt = pca.transform(Xt)
Yt /= np.sqrt((Yt**2).sum())
np.testing.assert_almost_equal(np.abs(Yt[0][0]), 1., 1)
def test_pca():
"""
PCA
"""
pca = PCA(n_comp=2)
X_r = pca.fit(X).transform(X)
np.testing.assert_equal(X_r.shape[1], 2)
pca = PCA()
pca.fit(X)
np.testing.assert_almost_equal(pca.explained_variance_ratio_.sum(),
1.0, 3)
def test_infer_dim_1():
"""
"""
n, p = 1000, 5
X = randn(n, p)*0.1 + randn(n, 1)*np.array([3, 4, 5, 1, 2]) + np.array(
[1, 0, 7, 4, 6])
pca = PCA(n_comp=p)
pca.fit(X)
spect = pca.explained_variance_
ll = []
for k in range(p):
ll.append(_assess_dimension_(spect, k, n, p))
ll = np.array(ll)
assert_true(ll[1] > ll.max() - .01 * n)
print __doc__
import pylab as pl
from scikits.learn import datasets
from scikits.learn.pca import PCA
from scikits.learn.lda import LDA
iris = datasets.load_iris()
X = iris.data
y = iris.target
target_names = iris.target_names
pca = PCA(n_components=2)
X_r = pca.fit(X).transform(X)
lda = LDA(n_components=2)
X_r2 = lda.fit(X, y).transform(X)
# Percentage of variance explained for each components
print 'explained variance ratio (first two components):', \
pca.explained_variance_ratio_
pl.figure()
pl.subplot(2, 1, 1)
for c, i, target_name in zip("rgb", [0, 1, 2], target_names):
pl.scatter(X_r[y == i, 0], X_r[y == i, 1], c=c, label=target_name)
pl.legend()
pl.title('PCA of IRIS dataset')
from scikits.learn.fastica import FastICA
if __name__ == '__main__':
###############################################################################
# Generate sample data
S = np.random.standard_t(1.5, size=(2, 10000))
S[0] *= 2.
# Mix data
A = [[1, 1], [0, 2]] # Mixing matrix
X = np.dot(A, S) # Generate observations
pca = PCA()
S_pca_ = pca.fit(X.T).transform(X.T).T
ica = FastICA()
S_ica_ = ica.fit(X).transform(X) # Estimate the sources
S_ica_ /= S_ica_.std(axis=1)[:,np.newaxis]
###############################################################################
# Plot results
def plot_samples(S, axis_list=None):
pl.scatter(S[0], S[1], s=2, marker='o', linewidths=0, zorder=10)
if axis_list is not None:
colors = [(0, 0.6, 0), (0.6, 0, 0)]
for color, axis in zip(colors, axis_list):
axis /= axis.std()
feature space) that account for the most variance in the data. Here we
plot the different samples on the 2 first principal components.
"""
print __doc__
import pylab as pl
from scikits.learn import datasets
from scikits.learn.pca import PCA
iris = datasets.load_iris()
X = iris.data
y = iris.target
target_names = iris.target_names
pca = PCA(n_components=2)
X_r = pca.fit(X).transform(X)
# Percentage of variance explained for each components
print pca.explained_variance_
pl.figure()
for c, i, target_name in zip("rgb", [0, 1, 2], target_names):
pl.scatter(X_r[y==i,0], X_r[y==i,1], c=c, label=target_name)
pl.legend()
pl.title('PCA of IRIS dataset')
pl.show()
import numpy as np
from scikits.learn.pca import PCA
from src.data_interface import d, L_clean, L
from src.utils import get_path, bool_to_color
path = get_path(__file__) + '/..'
L = list(L)
# Remove trial_id, obsnum and is alert
# I change notation here from D to X
X = d.view()[:,3:]
pca = PCA(n_components=30)
pca.fit(X)
plt.plot(np.cumsum(pca.explained_variance_ratio_), marker='o')
ax = plt.gca()
plt.title('Cumulative percentage of total variation explained by principal components')
ax.set_xlabel('Principal component')
ax.set_ylabel('% of total variation')
plt.savefig('{0}/plots/pca-variation-explained.pdf'.format(path), papertype='a4', format='pdf')
plt.cla()
W = pca.components_[:,0:3]
X_p = np.dot(X,W)
rnd_rows = np.random.random_integers(0, X.shape[0], 120)
colors = map(bool_to_color, d.view()[rnd_rows,L.index('IsAlert')])
from scikits.learn.pca import PCA
from scikits.learn.fastica import FastICA
###############################################################################
# Generate sample data
S = np.random.standard_t(1.5, size=(2, 10000))
S[0] *= 2.
# Mix data
A = [[1, 1], [0, 2]] # Mixing matrix
X = np.dot(A, S) # Generate observations
pca = PCA()
S_pca_ = pca.fit(X.T).transform(X.T).T
ica = FastICA()
S_ica_ = ica.fit(X).transform(X) # Estimate the sources
S_ica_ /= S_ica_.std(axis=1)[:, np.newaxis]
###############################################################################
# Plot results
def plot_samples(S, axis_list=None):
pl.scatter(S[0], S[1], s=2, marker='o', linewidths=0, zorder=10)
if axis_list is not None:
colors = [(0, 0.6, 0), (0.6, 0, 0)]
for color, axis in zip(colors, axis_list):
