Christian Widmer cwidmer

cwidmer / log_det.py

Last active August 29, 2015 14:07

different methods for log determinants

cwidmer / blocking_eigenvecs

Created October 1, 2014 19:53

Properties of blocking eigen decomposition

cwidmer / caching.py

Created March 20, 2014 23:58

	"""
	cache linear algebra (or any other) operations
	based on hashes of input arguments
	"""

	import numpy.linalg as la
	from sklearn.external import joblib

	mem = joblib.Memory(cachedir=cachedir, verbose=True, compress=True)
	#mem.clear()

cwidmer / index_timing.py

Last active August 29, 2015 13:57

	"""
	snippet to time different indexing strategies

	based on:
	http://wesmckinney.com/blog/?p=215
	http://stackoverflow.com/questions/11800075/faster-numpy-fancy-indexing-and-reduction/11813040#11813040
	http://stackoverflow.com/questions/14386822/fast-numpy-fancy-indexing?rq=1
	"""

	import numpy as np

cwidmer / pca_eigh.py

Last active August 29, 2015 13:56

perform pca via eigendecomposition

	def dual_pca(X, max_num_pcs):
	"""
	assuming N << D (X.shape[0] << X.shape[1]),
	we first compute the kernel matrix K = XX^T, perform an
	eigendecomposition and subsequently reconstruct
	truncated principle components
	"""


	K = X.dot(X.T)

cwidmer / experiment_toy_hmtmkl.py

Last active August 29, 2015 13:55

toy example for learning the similarity

	#!/usr/bin/env python2.6
	# This program is free software; you can redistribute it and/or modify
	# it under the terms of the GNU General Public License as published by
	# the Free Software Foundation; either version 2 of the License, or
	# (at your option) any later version.
	#
	# Written (W) 2013 Christian Widmer
	# Copyright (C) 2013 Max-Planck-Society, MSKCC, TU-Berlin

	"""

cwidmer / dendogram.py

Created January 30, 2014 09:15

plot matrix with dendogram

	import numpy as np
	import copy
	import random
	import pylab
	import scipy.cluster.hierarchy as sch


	def create_linkage_matrix(num_tasks):
	"""
	create stack of matrices for

cwidmer / shogun_string_factory.py

Created December 5, 2013 21:08

python module to create shogun objects for dealing with string kernels and string data withing the COFFIN framework

	#!/usr/bin/env python2.5
	# This program is free software; you can redistribute it and/or modify
	# it under the terms of the GNU General Public License as published by
	# the Free Software Foundation; either version 2 of the License, or
	# (at your option) any later version.
	#
	# Written (W) 2010-2013 Christian Widmer
	# Copyright (C) 2010-2013 Max-Planck-Society, TU-Berlin, MSKCC

	"""

cwidmer / viz_seq.py

Created December 5, 2013 20:49

Visualize a list of sequences (of same length, assuming DNA encoding) as a color coded matrix

	import pylab
	import numpy


	def show_seqs(seqs):
	"""
	plot color coded training sequences
	"""

cwidmer / coshuffle.py

Created December 5, 2013 20:46

Shuffle several iterables the same way

	import random


	def coshuffle(*args):
	"""
	will shuffle target_list and apply
	same permutation to other lists

	>>> helper.coshuffle([2, 1, 3], [4, 2, 8], [6, 3, 12])
	([5, 3, 2, 1, 4], [5, 3, 2, 1, 4], [5, 3, 2, 1, 4])