Dr.Q quinsulon

Principal Technologist/Engineer (SDE/Hands-on Architect) interested in creating mobile apps and data analyses for profit and public good ;]

quinsulon / From word2vec to doc2vec: similarity driven CRP, by Yingjie Miao.py

Created October 4, 2017 22:49 — forked from ryanpbrewster/From word2vec to doc2vec: similarity driven CRP, by Yingjie Miao.py

From word2vec to doc2vec --- similarity driven CRP by Yingjie Miao

	# vecs: an array of real vectors
	def crp(vecs):
	clusterVec = [] # tracks sum of vectors in a cluster
	clusterIdx = [] # array of index arrays. e.g. [[1, 3, 5], [2, 4, 6]]
	ncluster = 0
	# probablity to create a new table if new customer
	# is not strongly "similar" to any existing table
	pnew = 1.0/ (1 + ncluster)
	N = len(vecs)
	rands = random.rand(N) # N rand variables sampled from U(0, 1)

quinsulon / crp.py

Created October 4, 2017 22:50 — forked from kaustubhn/crp.py

Chinese Restaurant Process

	# Implementation of a chinese restaurant process function for a given list of word vectors.
	def crp(vecs):
	clusterVec = [[0.0] * 25] # tracks sum of vectors in a cluster
	clusterIdx = [[]] # array of index arrays. e.g. [[1, 3, 5], [2, 4, 6]]
	ncluster = 0
	# probablity to create a new table if new customer
	# is not strongly "similar" to any existing table
	pnew = 1.0/ (1 + ncluster)
	N = len(vecs)
	rands = [random.random() for x in range(N)] # N rand variables sampled from U(0, 1)