2011-02-02 18:36:10 +08:00
|
|
|
"""
|
|
|
|
|
Bench the scikit's ward implement compared to scipy's
|
|
|
|
|
"""
|
|
|
|
|
|
|
|
|
|
import time
|
|
|
|
|
|
|
|
|
|
import numpy as np
|
|
|
|
|
from scipy.cluster import hierarchy
|
|
|
|
|
import pylab as pl
|
|
|
|
|
|
2011-09-03 18:57:53 +08:00
|
|
|
from sklearn.cluster import Ward
|
2011-02-02 18:36:10 +08:00
|
|
|
|
2012-02-27 04:53:36 +08:00
|
|
|
ward = Ward(n_clusters=3)
|
2011-02-02 18:36:10 +08:00
|
|
|
|
2011-02-24 00:17:39 +08:00
|
|
|
n_samples = np.logspace(.5, 3, 9)
|
2011-02-02 18:36:10 +08:00
|
|
|
n_features = np.logspace(1, 3.5, 7)
|
2011-04-02 00:27:16 +08:00
|
|
|
N_samples, N_features = np.meshgrid(n_samples,
|
2011-02-02 18:36:10 +08:00
|
|
|
n_features)
|
|
|
|
|
scikits_time = np.zeros(N_samples.shape)
|
|
|
|
|
scipy_time = np.zeros(N_samples.shape)
|
|
|
|
|
|
|
|
|
|
for i, n in enumerate(n_samples):
|
|
|
|
|
for j, p in enumerate(n_features):
|
|
|
|
|
X = np.random.normal(size=(n, p))
|
|
|
|
|
t0 = time.time()
|
|
|
|
|
ward.fit(X)
|
|
|
|
|
scikits_time[j, i] = time.time() - t0
|
|
|
|
|
t0 = time.time()
|
2011-02-25 05:59:49 +08:00
|
|
|
hierarchy.ward(X)
|
2011-02-02 18:36:10 +08:00
|
|
|
scipy_time[j, i] = time.time() - t0
|
|
|
|
|
|
2011-12-17 03:18:40 +08:00
|
|
|
ratio = scikits_time / scipy_time
|
2011-02-02 18:36:10 +08:00
|
|
|
|
|
|
|
|
pl.clf()
|
|
|
|
|
pl.imshow(np.log(ratio), aspect='auto', origin="lower")
|
|
|
|
|
pl.colorbar()
|
|
|
|
|
pl.contour(ratio, levels=[1, ], colors='k')
|
|
|
|
|
pl.yticks(range(len(n_features)), n_features.astype(np.int))
|
|
|
|
|
pl.ylabel('N features')
|
|
|
|
|
pl.xticks(range(len(n_samples)), n_samples.astype(np.int))
|
|
|
|
|
pl.xlabel('N samples')
|
|
|
|
|
pl.title("Scikit's time, in units of scipy time (log)")
|
|
|
|
|
pl.show()
|
