13 Commits

Author	SHA1	Message	Date
rluna	72f2688c7d	finished of correcting "int" to "size_t" to avoid nasty errors and ... implement a test for SetWeights() function	2019-01-04 17:47:13 +01:00
rluna	f152b3030b	added posibility to change internal weights of the network directly ... assigning the values	2019-01-04 17:24:37 +01:00
rluna	e5eb1bbf9d	fix of some compiler warnings: comparison between unsigned int and int	2019-01-03 16:37:50 +01:00
David Nogueira	75116e157e	Update MLP.cpp	2018-09-20 13:02:56 +01:00
Artur Leonov	f783728532	add cmake build	2018-09-20 13:53:39 +03:00
davidjacnogueira	4f55a15543	Bug fix in Linux compilation.	2018-03-31 02:59:42 +01:00
davidjacnogueira	d44ec01e86	Edit README and Makefile.	2017-03-11 14:20:49 +00:00
davidjacnogueira	4005aff367	Add IRIS dataset usage example. ... Add LoadModel and SaveModel methods.	2017-03-10 00:20:35 +00:00
davidjacnogueira	0a636416ed	Internal architecture changes (to allow diferent activation functions for each layer and to allow hidden layers to have different number of nodes).	2016-11-09 02:31:53 +00:00
davidjacnogueira	f647b05f70	Add logging header file.	2016-11-03 23:05:54 +00:00
davidjacnogueira	9ff33f7b65	Slow learning time solved with equal number of positive and negative training samples.	2016-11-03 02:59:32 +00:00
davidjacnogueira	ff7bfe1fa2	Almost complete. ... Works with zero hidden layers, but not with one or more. Correctness in the backprop phase must be checked.	2016-11-02 01:04:34 +00:00
davidjacnogueira	a9e908bcbf	WIP MLP class still in development.	2016-01-22 02:15:36 +00:00