Hello! I’m trying to re-implement probabilistic matrix factorization (e.g. as described in section 2 in this paper ). Below is the gist of my code. tfidf is a (somewhat) large scipy.sparse.csr.csr_matrix, with entries that are positive and no larger than 1.0. Notice that I’m jumping through hoops wi…

find_MAP giving all zeros for sparse data

_eigenfoo August 31, 2018, 5:51am 7

Do you mean PyMC3’s implementation of find_MAP, or just MAP methods in general?

Topic		Replies	Views
StopIteration in find_map() Questions development , theano	14	1715	June 4, 2018
Unconsistent logp, Evaluating the results of find_MAP on Custom Likelihood Questions	3	610	August 28, 2018
Possible Performance Issue Using Sparse Matrices in Bipartite Noisy-Or Model Questions from_github , theano	10	2630	September 15, 2017
Nan while running find_MAP Questions	5	568	December 30, 2019
Best practices for MAP estimation Questions	3	1247	September 6, 2019