Vlad Niculae vene

I hereby claim:

To claim this, I am signing this object:

	# Author: Vlad Niculae <vlad@vene.ro>
	# License: 2-clause BSD

	"""2D implementation of the robust Siegel Repeated Median slope estimator

	This estimator tolerates corruption of up to 50% of the input points in either
	the X or the Y dimension.

	Vectorized implementation, and a naive implementation for sanity-check.
	"""

	/! Socket.IO.js build:0.9.16, development. Copyright(c) 2011 LearnBoost <dev@learnboost.com> MIT Licensed /

	/* Modifications by Vlad Niculae <vlad@vene.ro>

	Available at https://gist.github.com/vene/c0657d854ae74a4511d2

	Forked from https://raw.githubusercontent.com/Automattic/socket.io-client/ \
	0.9.16/dist/socket.io.js

	Changes:

	from collections import Counter
	import numpy as np
	from sklearn.metrics import euclidean_distances
	from pyemd import emd as pyemd


	def word_movers_distance(a, b, embeddings):
	"""Word Mover's Distance.

	A measure of text similarity: earth mover's distance in embedding metric space.

	""" Poisson-loss Factorization Machines with Numba

	Follows the vanilla FM model from:
	Steffen Rendle (2012): Factorization Machines with libFM.
	In: ACM Trans. Intell. Syst. Technol., 3(3), May.
	http://doi.acm.org/10.1145/2168752.2168771

	See also: https://github.com/coreylynch/pyFM
	"""

	# -- coding: utf8 --
	"""Convex factorization machines

	Implements the solver by: Mathieu Blondel, Akinori Fujino, Naonori Ueda.
	"Convex factorization machines". Proc. of ECML-PKDD 2015
	http://www.mblondel.org/publications/mblondel-ecmlpkdd2015.pdf
	"""

	# Author: Vlad Niculae <vlad@vene.ro>
	# License: Simplified BSD

	# low-rank bilinear regression using theano (supports sparse inputs)

	# predicts f(x_left, x_right) = x_left' UV' x_right
	# Reference:
	# Generalised Bilinear Regression
	# K. Ruben Gabriel
	# Source: Biometrika, Vol. 85, No. 3 (Sep., 1998), pp. 689-700
	# Stable URL: http://www.jstor.org/stable/2337396

	# Author: Vlad Niculae <vlad@vene.ro>