Sparse Matrix Representations of Feature Space using Eigen3

Co-occurrence analysis - bread and butter metrics and filtering provides the following tools:

samples2cooc - create co-occurrence matrix and feature list from samples
cooc2logl - compute log likelihood of co-occurrences
loglfilter - filter input co-occurrence matrix based on logl threshold
AB-cosine - compare feature vectors in co-occurrence matrix
show_features - human readable feature vectors using co-occurrence matrix and feature list

dumpmat - dump matrix contents to stdout useful for eyeballing computations and debugging

Name		Name	Last commit message	Last commit date
Latest commit History 39 Commits
cmake		cmake
.gitignore		.gitignore
AB_cosine.cpp		AB_cosine.cpp
CMakeLists.txt		CMakeLists.txt
CONTRIBUTING.md		CONTRIBUTING.md
README.md		README.md
TODO.md		TODO.md
cooc2logl.cpp		cooc2logl.cpp
dumpmat.cpp		dumpmat.cpp
feature_map.hpp		feature_map.hpp
index_bimap.hpp		index_bimap.hpp
loglfilter.cpp		loglfilter.cpp
mat2mtx.cpp		mat2mtx.cpp
samples2cooc.cpp		samples2cooc.cpp
show_features.cpp		show_features.cpp
showmat.cpp		showmat.cpp
sparse_matrix.hpp		sparse_matrix.hpp
sparse_matrix_io.hpp		sparse_matrix_io.hpp
testio.cpp		testio.cpp