Download presentation
Presentation is loading. Please wait.
1
libfann R binding 快速神经网络在R上的绑定
2
视神经传导分层 Picture from Simon. Thorpe
3
Nerual Network
4
Activation Function
5
libfann 介绍 主要特点： C开发、多层、易用、快速、多参数易调节
支持NN BP 多种方法 （RPROP, Quickprop, Batch, Incremental) 主要语言种类： C++、Java、Ada、PERL、PHP、Python、Ruby、Visual Prolog 7 计算相关的： Mathematica、Matlab、Octave
6
rfann下载及安装 获取软件包： https://sourceforge.net/projects/rfann/
获取软件包： *NIX 环境，gcc 开发工具环境 安装命令 R CMD INSTALL ./fann.tgz
7
rfann 之 R API fann_create fann_SetOpt fann_opts fann_Commit fann_train
fann_test fann_save fann_read fann_version_info
8
rfann参数 [1] " " [2] "network type [ LAYER | SHORTCUT ] Can't be set" [3] "network_type=FANN_NETTYPE_LAYER" [5] " " [6] "ann type [ Simple | Cascade | Expand | Sparse ] For Create" [9] " Basic parameters " [10] "init_weight=Init" [11] "num_layers=2" [12] "" [13] "num_input=2" [14] "num_output=1" [15] "num_neurons_hidden=3" [16] "desired_error= " [17] " Expand parameters " [18] "randomize_weight_high= " [19] "randomize_weight_low= " [20] "learning_rate= " [21] "connection_rate= " [22] "bit_fail_limit= " [23] "max_epochs=10000" [24] "epochs_between_reports=1000" [25] "learning_momentum= " [26] " " [27] "train algorithm: [ INCREMENTAL | BATCH | RPROP | QUICKPROP ]"
![rfann参数 [1] [2] network type [ LAYER | SHORTCUT ] Can t be set](http://slidesplayer.com/slide/16955527/97/images/8/rfann%E5%8F%82%E6%95%B0+%5B1%5D+%5B2%5D+network+type+%5B+LAYER+%7C+SHORTCUT+%5D+Can+t+be+set.jpg "[3] network_type=FANN_NETTYPE_LAYER [5] [6] ann type [ Simple | Cascade | Expand | Sparse ] For Create [9] Basic parameters [10] init_weight=Init [11] num_layers=2 [12] [13] num_input=2 [14] num_output=1 [15] num_neurons_hidden=3 [16] desired_error= [17] Expand parameters [18] randomize_weight_high= [19] randomize_weight_low= [20] learning_rate= [21] connection_rate= [22] bit_fail_limit= [23] max_epochs=10000 [24] epochs_between_reports=1000 [25] learning_momentum= [26] [27] train algorithm: [ INCREMENTAL | BATCH | RPROP | QUICKPROP ]")
9
rfann参数 cont. [30] " " [31] "error function: [ LINEAR | TANH ]" [32] "train_error_function=FANN_ERRORFUNC_TANH" [33] "" [34] " " [35] "stop function: [ MSE | BIT ]" [36] "train_stop_function=FANN_STOPFUNC_BIT" [37] "" [38] " rprop parameters " [39] "quickprop_decay= " [40] "quickprop_mu= " [41] "rprop_increase_factor= " [42] "rprop_decrease_factor= " [43] "rprop_delta_min= " [44] "rprop_delta_max= " [45] "rprop_delta_zero= " [46] "" [47] "activation_steepness_hidden= " [48] "activation_steepness_output= " [49] " " [50] "Options below [ LINEAR | THRESHOLD | THRESHOLD_SYMMETRIC ]" [51] "cont: [ SIGMOID| SIGMOID_STEPWISE| SIGMOID_SYMMETRIC | SIGMOID_SYMMETRIC_STEPWISE ]" [52] "cont: [ GAUSSIAN| GAUSSIAN_SYMMETRIC | GAUSSIAN_STEPWISE ]" [53] "cont: [ ELLIOT | ELLIOT_SYMMETRIC" [54] "cont: [ LINEAR_PIECE | LINEAR_PIECE_SYMMETRIC ]" [55] "cont: [ SIN_SYMMETRIC| COS_SYMMETRIC | SIN | COS ]" [56] " " [57] "activation_function_hidden=FANN_SIGMOID_SYMMETRIC" [58] "activation_function_output=FANN_GAUSSIAN_SYMMETRIC"
![rfann参数 cont. [30] [31] error function: [ LINEAR | TANH ]](http://slidesplayer.com/slide/16955527/97/images/9/rfann%E5%8F%82%E6%95%B0+cont.+%5B30%5D+%5B31%5D+error+function%3A+%5B+LINEAR+%7C+TANH+%5D.jpg "[32] train_error_function=FANN_ERRORFUNC_TANH [33] [34] [35] stop function: [ MSE | BIT ] [36] train_stop_function=FANN_STOPFUNC_BIT [37] [38] rprop parameters [39] quickprop_decay= [40] quickprop_mu= [41] rprop_increase_factor= [42] rprop_decrease_factor= [43] rprop_delta_min= [44] rprop_delta_max= [45] rprop_delta_zero= [46] [47] activation_steepness_hidden= [48] activation_steepness_output= [49] [50] Options below [ LINEAR | THRESHOLD | THRESHOLD_SYMMETRIC ] [51] cont: [ SIGMOID| SIGMOID_STEPWISE| SIGMOID_SYMMETRIC | SIGMOID_SYMMETRIC_STEPWISE ] [52] cont: [ GAUSSIAN| GAUSSIAN_SYMMETRIC | GAUSSIAN_STEPWISE ] [53] cont: [ ELLIOT | ELLIOT_SYMMETRIC [54] cont: [ LINEAR_PIECE | LINEAR_PIECE_SYMMETRIC ] [55] cont: [ SIN_SYMMETRIC| COS_SYMMETRIC | SIN | COS ] [56] [57] activation_function_hidden=FANN_SIGMOID_SYMMETRIC [58] activation_function_output=FANN_GAUSSIAN_SYMMETRIC")
10
例程 library(fann) x<-c( 1, 1,-1,-1, 1,-1, 1,-1) dim(x) <- c(4,2)
y<-c(-1,1,1,-1) ann<-fann_create(num_layers=3,layer1=2,layer2=2,layer3=1,num_neurons_hidden=3); fann_SetOpt(ann,bit_fail_limit=0.35,train_stop_function="MSE") fann_SetOpt(ann, ) fann_Commit(ann) fann_train(ann,x,y) fann_save(ann,"./xortemp.net") o<-fann_test(ann,x,y)
11
rfann中的结构 Nncomon.h FannOpt.c FannProc.c FannData.c
typedef struct FANNOPT { char *optname; int type; FANNCODE opt; // (void *) value; } FannOpt; FannOpt.c FannProc.c FannData.c
12
遗留问题 *NIX 版本 list 类型的支持 fann GUI 目前在R-exts还不能解析list类型 图形化显示rfann结构
图形化显示训练过程
13
rfann适用局限性 训练时间长：人工神经网络需要长时间的训练，有时可能使之变得不实用。大多数简单问题的网络训练需要至少上千次迭代，复杂问题的训练可能需要多达数万次迭代。 需大量训练数据：只有少量输入-输出数据，一般不考虑使用人工神经网络。 不能保证最佳结果：训练可能导致网络发生偏离，使之在一些操作区域内结果准确，而在其他区域则不准确。此外，在训练过程中，有可能偶尔陷入“局部最小”。 不能保证完全可靠：尽管这一点对所有的计算问题均适用，但对人工神经网络尤其如此。需要高度可靠的问题，在采用人工神经网络时必须小心谨慎。生物制药【Eric Xing】、金融程序交易④、火控；
14
结论：神经网络的隐层 隐层数 通常增加隐层数可以降低网络误差（也有文献认为不一定能有效降低），提高精度，但也使网络复杂化，从而增加了网络的训练时间和出现“过拟合”的倾向。Hornik等早已证明：若输入层和输出层采用线性转换函数，隐层采用Sigmoid转换函数，则含一个隐层的MLP网络能够以任意精度逼近任何有理函数。□ 一般地，靠增加隐层节点数来获得较低的误差，其训练效果要比增加隐层数更容易实现。
15
推荐阅读材料 Steffen Nissen Implementation of a Fast Artificial Neural Network Library (fann) Anil K.Jain Jianchang Mao K.M Mohiuddin: Artificial Neural Network:: A Tutorial IEEE March 1996 Ben Krose Patrick van der Smagt An introduction to Neural Networks eighth edition JingTao YAO &Chew Lim TAN （ ） Guidelines for Financial Forecasting with Neural Networks
16
感谢
Similar presentations
