Apologies for the delay; attached are brief summaries of my PDL modules
on CPAN -- looks like your scan didn't turn up all of them (probably due
to CPAN's failure to correctly handle PODs generated by PDL::PP, as
David pointed out to me a while back)...
happy merrry marmosets,
Bryan
--
Bryan Jurish "There is *always* one more bug."
[email protected] -Lubarsky's Law of Cybernetic Entomology
=head1 PDL::VectorValued
PDL::VectorValued - Utilities for vector-valued PDLs
PDL::VectorValued provides generalizations of some elementary PDL
functions to higher-order PDLs which treat vectors as "data values",
as well as some PDL::PP macro-like
routines for building derivative modules using binary search of sorted
vector-valued PDLs.
=head1 PDL::Ngrams
PDL::Ngrams - N-Gram utilities for PDL
PDL::Ngrams provides basic utilities for tracking N-grams over PDL vectors.
In particular includes utilities for co-occurrence matrix construction,
insertion
and removal of subsequence delimiters.
=head1 PDL::CCS
PDL::CCS - Sparse N-dimensional PDLs
PDL::CCS::Nd is a Perl class for representing large sparse N-dimensional
numeric structures using sorted index vector-vectors and a flat vector of
non-missing values, which supports a subset of the perl-side PDL API.
The PDL::CCS distribution also includes some low-level generic PDL::PP
utilities for Harwell-Boeing encoding and decoding
"pointers" along arbitrary dimensions of a sparse PDL given an index list,
as well as utilities for block-wise alignment of pairs of sparse
index-encoded PDLs. Requires PDL::VectorValued.
=head1 PDL::HMM
PDL::HMM - Discrete Hidden Markov Model implementation for PDL
PDL::HMM provides an implementation of discrete 1st-order
Hidden Markov Models encoded as PDLs. Includes code for
sequence probability computation (foward- and backward-algorithms),
sequence decoding (Viterbi algorithm), and parameter estimation
(forward-backward aka Baum-Welch algorithm). PDL::Ngrams can be used
to produce maximum-likelihood transition models from appropriate
training sequences.
=head1 PDL::GA
PDL::GA - Genetic algorithm utilities for PDLs
Some very basic utilities for implementing genetic algorithms on PDLs.
Includes some routines for weighted and unweighted selection, bitwise
coding, mutation, and crossover.
=head1 PDL::EditDistance
PDL::EditDistance - Parameterized string edit distance and alignment for PDLs
PDL::EditDistance provides PDL routines for computing string edit distances
and alignment matrices for pairs of PDL vectors using the parameterized
Fischer-Wagner algorithm.
=head1 PDL::SVDLIBC
PDL::SVDLIBC - PDL interface to Doug Rohde's SVD C Library
PDL::SVDLIBC provides a PDL interface to the SVDLIBC routines
for singular value decomposition of large sparse matrices using
the Lanczos approximation. PDL::CCS can be used to create the
Harwell-Boeing encoded format from a vanilla PDL.
_______________________________________________
Perldl mailing list
[email protected]
http://mailman.jach.hawaii.edu/mailman/listinfo/perldl
