神经网络其实可以处理很大的数据,也可以处理音频。接下来我们就用使用上次的栗子来验证一次,但这次的目标输出是一个2983*1的音频,远远大于之前的10*1。
同样的还是一步一步的来
Step1 :Prepare Data
我们还是使用MNIST_small作为输入。(MNIST_small的介绍请参照上一篇博客)
但这次,我们不使用现有的标签作为目标输出,而是使用已经录制好的10个数字的音频(2983*10)作为神经网络的目标输出。(音频及代码框架下载地址)
代码如下:
% prepare the data set
load mnist_small_matlab.mat;
train_size = 10000;
X_train{1} = reshape(trainData(1: 14, 1: 14, :), [], train_size);
X_train{2} = reshape(trainData(15: 28, 1: 14, :), [], train_size);
X_train{3} = reshape(trainData(15: 28, 15: 28, :), [], train_size);
X_train{4} = reshape(trainData(1: 14, 15: 28, :), [], train_size);
X_train{5} = zeros (0, train_size);
X_train{6} = zeros (0, train_size);
X_train{7} = zeros (0, train_size);
X_train{8} = zeros (0, train_size);
test_size = 2000;
X_test{1} = reshape(testData(1: 14, 1: 14, :), [], test_size);
X_test{2} = reshape(testData(15: 28, 1: 14, :), [], test_size);
X_test{3} = reshape(testData(15: 28, 15: 28, :), [], test_size);
X_test{4} = reshape(testData(1: 14, 15: 28, :), [], test_size);
X_test{5} = zeros (0, test_size);
X_test{6} = zeros (0, test_size);
X_test{7} = zeros (0, test_size);
X_test{8} = zeros (0, test_size);
% prepare standard speech audio
audio_list = dir('audio/*.wav');
for i = 1 : numel(audio_list)
audio_name = audio_list(i).name;
[aud, ~] = audioread(strcat('audio/', audio_name));
end
目标输出的音频处理放在之后介绍
Step2 :Design Network Architecture
代码如下:
% define network architecture
layer_size = [196 6000
196 6000
196 6000
196 6000
0 4000
0 4000
0 4000
0 2983];
L = 8;
Step3 : Initialize Parameters
Initialize Weights
for l = 1: L - 1
w{l} = (rand(layer_size(l + 1, 2), sum(layer_size(l, :))) * 2 - 1) * sqrt(6 / (layer_size(l + 1, 2) + sum(layer_size(l, :))));
end
Choose Parameters
alpha = 1; %learning rate 学习率
max_iter = 300; %number of iteration 迭代次数
mini_batch = 100; %number of samples in a batch 每一批处理的样本个数
注:mini_batch表示每一次批处理的样本个数,即每次批处理100个样本,而不是直接处理10000个样本,每次处理的100个样本是随机从10000个样本中选择出来的。
Step4 : Run the Network
激活函数
经验表明,以ReLU函数作为激活函数往往能够取得较好的训练效果。在本次试验中,除倒数第二层外,其余层均使用ReLU函数作为激活函数。
神经网络的输出是一个有10个元素的列向量,这个列向量只能有一位为1,其余为0,第几位为1表示这是数字几。如第0位为1,则判断该数字为0.(从0开始数数)
考虑到神经网络的输出,我们在最后一层的前一层使用sigmoid函数作为激活函数,以保证输出的结果为0到1直接的数。举个栗子,以数字‘8’作为输入,若输出的结果向量中的第八位非常接近1,其余位接近0,则认为该样本为数字‘8’,神经网络的输出结果正确。
前向计算
ReLU函数:
function [a_next, z_next] = fc(w, a, x)
% define the activation function
f = @(s) max(0, s);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Your code BELOW
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% forward computing (either component or vector form)
a = [x
a];
z_next = w * a;
a_next = f(z_next);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Your code ABOVE
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
end
Sigmoid函数:
function [a_next, z_next] = fc2(w, a, x)
% define the activation function
f = @(s) 1 ./ (1 + exp(-s));
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Your code BELOW
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% forward computing (either component or vector form)
a = [x
a];
z_next = w * a;
a_next = f(z_next);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Your code ABOVE
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
end
Cost Function
J = [J 1/2/mini_batch*sum((a{L}(:) - y(:)).^2)];
后向计算
ReLU函数:
function delta = bc(w, z, delta_next)
% define the activation function
f = @(s) max(0, s);
% define the derivative of activation function
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Your code BELOW
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% backward computing (either component or vector form)
xxj = size(z, 1);
delta = w' * delta_next;
df = [];
for i = 1 : size(z, 1)
for j = 1 : size(z, 2)
if z(i, j) > 0
df(i, j) = 1;
else
df(i, j) = 0;
end
end
end
delta = delta(1 : xxj, :) .* df;
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Your code ABOVE
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
end
Sigmoid函数:
function delta = bc2(w, z, delta_next)
% define the activation function
f = @(s) 1 ./ (1 + exp(-s));
% define the derivative of activation function
df = @(s) f(s) .* (1 - f(s));
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Your code BELOW
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% backward computing (either component or vector form)
xxj = size(z, 1);
delta = w' * delta_next;
delta = delta(1 : xxj, :) .* df(z);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Your code ABOVE
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
end
更新权值
for l = 1 : L - 1
gw = delta{l + 1} * [x{l}; a{l}]' / mini_batch;
w{l} = w{l} - alpha * gw;
end
Step5 : Evaluation
Acc = number of correct predictionsnumber of samples
Accuracy of training set :
a{1} = zeros(layer_size(1, 2), train_size);
for l = 1 : L - 1
a{l + 1} = fc(w{l}, a{l}, X_train{l});
end
[~, ind_train] = max(trainLabels);
[~, ind_pred] = max(a{L});
train_acc = sum(ind_train == ind_pred) / train_size;
fprintf('Accuracy on training dataset is %f%%\n', train_acc * 100);
Accuracy of testing set :
a{1} = zeros(layer_size(1, 2), test_size);
for l = 1 : L - 1
a{l + 1} = fc(w{l}, a{l}, X_test{l});
end
[~, ind_test] = max(testLabels);
[~, ind_pred] = max(a{L});
test_acc = sum(ind_test == ind_pred) / test_size;
fprintf('Accuracy on testing dataset is %f%%\n', test_acc * 100);
完整代码:(目标输出的处理请参照代码)
% clear workspace and close plot windows
clear;
close all;
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Your code BELOW
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% prepare the data set
load mnist_small_matlab.mat;
train_size = 10000;
X_train{1} = reshape(trainData(1: 14, 1: 14, :), [], train_size);
X_train{2} = reshape(trainData(15: 28, 1: 14, :), [], train_size);
X_train{3} = reshape(trainData(15: 28, 15: 28, :), [], train_size);
X_train{4} = reshape(trainData(1: 14, 15: 28, :), [], train_size);
X_train{5} = zeros (0, train_size);
X_train{6} = zeros (0, train_size);
X_train{7} = zeros (0, train_size);
X_train{8} = zeros (0, train_size);
test_size = 2000;
X_test{1} = reshape(testData(1: 14, 1: 14, :), [], test_size);
X_test{2} = reshape(testData(15: 28, 1: 14, :), [], test_size);
X_test{3} = reshape(testData(15: 28, 15: 28, :), [], test_size);
X_test{4} = reshape(testData(1: 14, 15: 28, :), [], test_size);
X_test{5} = zeros (0, test_size);
X_test{6} = zeros (0, test_size);
X_test{7} = zeros (0, test_size);
X_test{8} = zeros (0, test_size);
% prepare standard speech audio
audio_list = dir('audio/*.wav');
for i = 1 : numel(audio_list)
audio_name = audio_list(i).name;
[aud, ~] = audioread(strcat('audio/', audio_name));
end
% choose parameters
alpha = 1;
max_iter = 300;
mini_batch = 100;
J = [];
Acc = [];
% define network architecture
layer_size = [196 6000
196 6000
196 6000
196 6000
0 4000
0 4000
0 4000
0 2983];
L = 8;
% initialize weights
for l = 1: L - 1
w{l} = (rand(layer_size(l + 1, 2), sum(layer_size(l, :))) * 2 - 1) * sqrt(6 / (layer_size(l + 1, 2) + sum(layer_size(l, :))));
end
% train
for iter = 1 : max_iter
ind = randperm(train_size);
for k = 1 : ceil(train_size / mini_batch)
a{1} = zeros(layer_size(1, 2), mini_batch);
for l = 1 : L
x{l} = X_train{l}(:, ind((k - 1) * mini_batch + 1 : min(k * mini_batch, train_size)));
end
% 目标输出
y = [];
y2 = [];
y1 = double(trainLabels( :, ind((k - 1) * mini_batch + 1 : min(k * mini_batch, train_size))));
for i = 1 : 100
y2 = find(y1(:, i)) - 1;
if y2 ~= 0
audio_name = audio_list(y2).name;
else
ausio_name = 0;
end
y = [y double(audioread(strcat('audio/', audio_name)))];
end
for l = 1 : L-2
[a{l + 1}, z{l + 1}] = fc(w{l}, a{l}, x{l});
end
[a{L}, z{L}] = fc2(w{L - 1}, a{L - 1}, x{L - 1});
J = [J 1/2/mini_batch*sum((a{L}(:) - y(:)).^2)];
fprintf('J=%.4f\n', J);
[~, ind_y] = max(y);
[~, ind_pred] = max(a{L});
xxj = sum(ind_y == ind_pred) / mini_batch;
Acc = [Acc xxj];
delta{L} = (a{L} - y) .* a{L} .* (1 - a{L});
delta{L - 1} = bc2(w{L - 1}, z{L - 1}, delta{L});
for l = L - 2 : -1 : 2
delta{l} = bc(w{l}, z{l}, delta{l + 1});
end
for l = 1 : L - 1
gw = delta{l + 1} * [x{l}; a{l}]' / mini_batch;
w{l} = w{l} - alpha * gw;
end
end
end
figure
plot(J);
figure
plot(Acc);
% save model
save model.mat w layer_size
% display/listen to some results pairs
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Your code ABOVE
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
我的结果太差了。。。还是不贴结果了
不知道是不是神经网络的设计又问题
求指导!