AI-NeuralNet-FastSOM
view release on metacpan or search on metacpan
- now using Drand01() instead of rand()
- now using seedDrand01() instead of srand()
- fixed problem with not using all training vectors, or some twice
- removed non-core Data::Dumper from tests
- added tests for store/retrieve via Storable
- first public release
0.06 Wed Jul 22 12:07:25 2009
- removed AI::NN::FSOM::ARRAY, ::MAP, and ::VECTOR modules
- removed Inline::C code from remaining modules
- removed dependence on non-core parent.pm
- removed remaining Inline::C macros and INLINE.h
- moved train() into C
- now parsing input_ and output_dim parameters (finally!)
0.05 Mon Jul 20 13:20:06 2009
- re-added support for labels, originally in AI::NN::SOM
- added original AI::NN::SOM test suite (and it works!)
0.04 Sat Jul 18 16:45:27 2009
- removed dependence on Inline::C
*
* Each of Rect, Map, Array, and Vector contains a member 'ref' which is
* an SV* pointing to an RV. The RV can be returned directly to perl-land
* after being blessed into its respective class.
*
* The RV references an SV containing an IV. The IV is set to the base
* address of its component structure. This is so the class code can know
* which instance of the class is being referred to on callback.
*
* The reference count of the SV has its initial reference count set to one,
* representing its parents ownership. If a parent dies or a perl-land
* reference is taken of any componenet, its reference count should
* be adjusted accordingly.
*
* When the count reaches zero perl will call the classes DESTROY method,
* at which point we can decrease the reference count on each child and
* free the component structure.
*
* The intent of all this reference count tom-foolery is to keep the
* component structures from disappearing from underneath perl-land
* references to them. As a bonus, we get a neat destruction mechanism
( run in 0.479 second using v1.01-cache-2.11-cpan-a5abf4f5562 )