Gael Varoquaux GaelVaroquaux

Research director at Inria. Computer science, data science, health. Scipy & pydata coder; (co)funder of scikit-learn, joblib, probabl.

GaelVaroquaux / sklearn_EN_example.py

Created September 29, 2012 16:16 — forked from arokem/sklearn_EN_example.py

Demonstrate that ElasticNet doesn't fit all the way, even for rather strict convergence criteria

	import numpy as np
	import sklearn.linear_model as lm

	X = np.array([[ -2.18252949e-01, -8.21949578e-02, -4.64055457e-02,
	-1.78405908e-01, -1.93863740e-01, 5.30667625e-02,
	1.83851107e-01, 1.23426449e-01, 1.97396315e-01,
	-2.12615837e-01, 7.06452283e-02, -1.94509405e-01,
	-9.77929516e-02, 2.07135018e-01, -3.40368338e-02,
	2.02970673e-01, -2.28669466e-01, 4.17398420e-02,
	1.80163132e-01, 3.24254938e-02, -2.41198452e-03,

GaelVaroquaux / lasso.py

Created October 5, 2012 07:12 — forked from aweinstein/lasso.py

scikitlearn lasso path fat vs thin X matrix

	import numpy as np
	import matplotlib.pyplot as plt
	from sklearn.linear_model import Lasso, lars_path

	np.random.seed(42)

	def gen_data(n, m, k):
	X = np.random.randn(n, m)
	w = np.zeros((m, 1))
	i = np.arange(0, m)

GaelVaroquaux / strategies_comparison.py

Last active May 17, 2016 20:22 — forked from aabadie/strategies_comparison.py

Persistence strategies comparison


	"""Persistence strategies comparison script.

	This script compute the speed, memory used and disk space used when dumping and
	loading arbitrary data. The data are taken among:
	- scikit-learn Labeled Faces in the Wild dataset (LFW)
	- a fully random numpy array with 10000x10000 shape
	- a dictionary with 1M random keys/values
	- a list containing 10M random value
	The compared persistence strategies are: