netpack_snn.m

来自「神经网络的工具箱, 神经网络的工具箱,」· M 代码 · 共 71 行

function netpack_snn()
%NETPACK_SNN SNN netpack toolbox.
% -----------------------------------------------------------------------
%
% The SNN netpack toolbox. 
%
% This toolbox can be used to train a feed forward neural network.
%
% -----------------------------------------------------------------------
% -----------------------------------------------------------------------
%
% FUNCTIONALITY 
% =============
%
% Supported training algorithms:
%   TRAINGD_SNN  - Gradient Descent
%   TRAINCGP_SNN - Conjugate Gradient (Polak Ribiere)
%   TRAINGLM_SNN - Levenberg-Marquardt
% these algorithms minimize a cost function.
%
% Supported cost functions:
%
%   WCF_SNN 
%     - E(y, t) = sum_{ \{i, \mu | \Delta_{i\mu} = 1\} } 
%                   a_i g_{\mu} f^i(y_{i\mu}, t_{i\mu})
%
%     with:
%       i        - output index
%       \mu      - pattern index
%       a_i      - output weight
%       g_{\mu}  - pattern weight
%       f^i - error function for output i
%       y_{i\mu} - prediction for output i, pattern \mu.
%       t_{i\mu} - target for output i, pattern \mu.
%       \Delta_{i\mu} - parameter controlling over which outputs for
%                       which patterns the summation is done
%
% Supported output error functions:
%
%   SE_SNN             - squared error
%   RELERR_SNN         - relative error
%   CROSSENTROPY_SNN   - cross entropy error 
%   CROSSLOGISTIC_SNN  - cross logistic error
%   LOGLIKELIHOOD_SNN  - loglikelihood error
%
% -----------------------------------------------------------------------
% 
% FUNCTIONS OVERVIEW
% ==================
%
% Network initialization: NET_STRUCT_SNN
% +-- Cost functions: WCF_SNN 
% |   +-- Output Error functions: SE_SNN, RELERR_SNN, CROSSENTROPY_SNN, 
% |   |                           CROSSLOGISTIC_SNN, LOGLIKELIHOOD_SNN
% +-- Training algorithms: TRAINGD_SNN, TRAINCGP_SNN, TRAINLM_SNN
% +-- Transfer functions: LINTF_SNN, EXPTF_SNN, LOGSIGTF_SNN,
%                         RADBASTF_SNN, TANSIGTF_SNN
%
% Data initialization: IMPORT_ASCII_DATA_SNN, WCFDATA_STRUCT_SNN
%
% Training functions: TRAIN_SNN, TRAIN_BOOTSTRAP_SNN, TRAIN_HALFOUT_SNN
%
% Ensemble functions: BALANCE_SNN, SIMFF_AVR_SNN 
%
% Confidence/prediction: C_CONFIDENCE_SNN, CONFIDENCE_SNN,
%                        C_PREDICTION_SNN, PREDICTION_SNN
%
%
% -----------------------------------------------------------------------
%