&>&Evies面板数据分析软件
Evies面板数据分析软件
上传大小：8.87MB
用于面板数据分析，可进行固定效应和随机效应的分析
综合评分：0
下载个数：
{%username%}回复{%com_username%}{%time%}\
/*点击出现回复框*/
$(".respond_btn").on("click", function (e) {
$(this).parents(".rightLi").children(".respond_box").show();
e.stopPropagation();
$(".cancel_res").on("click", function (e) {
$(this).parents(".res_b").siblings(".res_area").val("");
$(this).parents(".respond_box").hide();
e.stopPropagation();
/*删除评论*/
$(".del_comment_c").on("click", function (e) {
var id = $(e.target).attr("id");
$.getJSON('/index.php/comment/do_invalid/' + id,
function (data) {
if (data.succ == 1) {
$(e.target).parents(".conLi").remove();
alert(data.msg);
$(".res_btn").click(function (e) {
var parentWrap = $(this).parents(".respond_box"),
q = parentWrap.find(".form1").serializeArray(),
resStr = $.trim(parentWrap.find(".res_area_r").val());
console.log(q);
//var res_area_r = $.trim($(".res_area_r").val());
if (resStr == '') {
$(".res_text").css({color: "red"});
$.post("/index.php/comment/do_comment_reply/", q,
function (data) {
if (data.succ == 1) {
var $target,
evt = e || window.
$target = $(evt.target || evt.srcElement);
var $dd = $target.parents('dd');
var $wrapReply = $dd.find('.respond_box');
console.log($wrapReply);
//var mess = $(".res_area_r").val();
var mess = resS
var str = str.replace(/{%header%}/g, data.header)
.replace(/{%href%}/g, 'http://' + window.location.host + '/user/' + data.username)
.replace(/{%username%}/g, data.username)
.replace(/{%com_username%}/g, data.com_username)
.replace(/{%time%}/g, data.time)
.replace(/{%id%}/g, data.id)
.replace(/{%mess%}/g, mess);
$dd.after(str);
$(".respond_box").hide();
$(".res_area_r").val("");
$(".res_area").val("");
$wrapReply.hide();
alert(data.msg);
}, "json");
/*删除回复*/
$(".rightLi").on("click", '.del_comment_r', function (e) {
var id = $(e.target).attr("id");
$.getJSON('/index.php/comment/do_comment_del/' + id,
function (data) {
if (data.succ == 1) {
$(e.target).parent().parent().parent().parent().parent().remove();
$(e.target).parents('.res_list').remove()
alert(data.msg);
//填充回复
function KeyP(v) {
var parentWrap = $(v).parents(".respond_box");
parentWrap.find(".res_area_r").val($.trim(parentWrap.find(".res_area").val()));
评论共有0条
VIP会员动态
CSDN下载频道资源及相关规则调整公告V11.10
下载频道用户反馈专区
下载频道积分规则调整V1710.18
spring mvc+mybatis+mysql+maven+bootstrap 整合实现增删查改简单实例.zip
资源所需积分/C币
当前拥有积分
当前拥有C币
输入下载码
为了良好体验，不建议使用迅雷下载
Evies面板数据分析软件
会员到期时间：
剩余下载个数：
剩余积分：0
为了良好体验，不建议使用迅雷下载
积分不足！
资源所需积分/C币
当前拥有积分
您可以选择
程序员的必选
绿色安全资源
资源所需积分/C币
当前拥有积分
当前拥有C币
为了良好体验，不建议使用迅雷下载
资源所需积分/C币
当前拥有积分
当前拥有C币
为了良好体验，不建议使用迅雷下载
资源所需积分/C币
当前拥有积分
当前拥有C币
您的积分不足，将扣除 10 C币
为了良好体验，不建议使用迅雷下载
无法举报自己的资源
你当前的下载分为234。
你还不是VIP会员
开通VIP会员权限，免积分下载
你下载资源过于频繁，请输入验证码
您因违反CSDN下载频道规则而被锁定帐户，如有疑问，请联络:!
若举报审核通过，可返还被扣除的积分
被举报人：
raohuaxiang
举报的资源分：
请选择类型
资源无法下载 （ 404页面、下载失败、资源本身问题）
资源无法使用 （文件损坏、内容缺失、题文不符）
侵犯版权资源 （侵犯公司或个人版权）
虚假资源 （恶意欺诈、刷分资源）
含色情、危害国家安全内容
含广告、木马病毒资源
*详细原因：
Evies面板数据分析软件&>&面板数据分析的SAS实现
面板数据分析的SAS实现
上传大小：134KB
面板数据的一般线性模型 固定效应模型的估计方法 随机效应模型的估计方法
综合评分：3
下载个数：
{%username%}回复{%com_username%}{%time%}\
/*点击出现回复框*/
$(".respond_btn").on("click", function (e) {
$(this).parents(".rightLi").children(".respond_box").show();
e.stopPropagation();
$(".cancel_res").on("click", function (e) {
$(this).parents(".res_b").siblings(".res_area").val("");
$(this).parents(".respond_box").hide();
e.stopPropagation();
/*删除评论*/
$(".del_comment_c").on("click", function (e) {
var id = $(e.target).attr("id");
$.getJSON('/index.php/comment/do_invalid/' + id,
function (data) {
if (data.succ == 1) {
$(e.target).parents(".conLi").remove();
alert(data.msg);
$(".res_btn").click(function (e) {
var parentWrap = $(this).parents(".respond_box"),
q = parentWrap.find(".form1").serializeArray(),
resStr = $.trim(parentWrap.find(".res_area_r").val());
console.log(q);
//var res_area_r = $.trim($(".res_area_r").val());
if (resStr == '') {
$(".res_text").css({color: "red"});
$.post("/index.php/comment/do_comment_reply/", q,
function (data) {
if (data.succ == 1) {
var $target,
evt = e || window.
$target = $(evt.target || evt.srcElement);
var $dd = $target.parents('dd');
var $wrapReply = $dd.find('.respond_box');
console.log($wrapReply);
//var mess = $(".res_area_r").val();
var mess = resS
var str = str.replace(/{%header%}/g, data.header)
.replace(/{%href%}/g, 'http://' + window.location.host + '/user/' + data.username)
.replace(/{%username%}/g, data.username)
.replace(/{%com_username%}/g, data.com_username)
.replace(/{%time%}/g, data.time)
.replace(/{%id%}/g, data.id)
.replace(/{%mess%}/g, mess);
$dd.after(str);
$(".respond_box").hide();
$(".res_area_r").val("");
$(".res_area").val("");
$wrapReply.hide();
alert(data.msg);
}, "json");
/*删除回复*/
$(".rightLi").on("click", '.del_comment_r', function (e) {
var id = $(e.target).attr("id");
$.getJSON('/index.php/comment/do_comment_del/' + id,
function (data) {
if (data.succ == 1) {
$(e.target).parent().parent().parent().parent().parent().remove();
$(e.target).parents('.res_list').remove()
alert(data.msg);
//填充回复
function KeyP(v) {
var parentWrap = $(v).parents(".respond_box");
parentWrap.find(".res_area_r").val($.trim(parentWrap.find(".res_area").val()));
评论共有1条
ppt，了解大部分的模型
VIP会员动态
CSDN下载频道资源及相关规则调整公告V11.10
下载频道用户反馈专区
下载频道积分规则调整V1710.18
spring mvc+mybatis+mysql+maven+bootstrap 整合实现增删查改简单实例.zip
资源所需积分/C币
当前拥有积分
当前拥有C币
输入下载码
为了良好体验，不建议使用迅雷下载
面板数据分析的SAS实现
会员到期时间：
剩余下载个数：
剩余积分：0
为了良好体验，不建议使用迅雷下载
积分不足！
资源所需积分/C币
当前拥有积分
您可以选择
程序员的必选
绿色安全资源
资源所需积分/C币
当前拥有积分
当前拥有C币
为了良好体验，不建议使用迅雷下载
资源所需积分/C币
当前拥有积分
当前拥有C币
为了良好体验，不建议使用迅雷下载
资源所需积分/C币
当前拥有积分
当前拥有C币
您的积分不足，将扣除 10 C币
为了良好体验，不建议使用迅雷下载
无法举报自己的资源
你当前的下载分为234。
你还不是VIP会员
开通VIP会员权限，免积分下载
你下载资源过于频繁，请输入验证码
您因违反CSDN下载频道规则而被锁定帐户，如有疑问，请联络:!
若举报审核通过，可返还被扣除的积分
被举报人：
zhaobaojiang
举报的资源分：
请选择类型
资源无法下载 （ 404页面、下载失败、资源本身问题）
资源无法使用 （文件损坏、内容缺失、题文不符）
侵犯版权资源 （侵犯公司或个人版权）
虚假资源 （恶意欺诈、刷分资源）
含色情、危害国家安全内容
含广告、木马病毒资源
*详细原因：
面板数据分析的SAS实现配置数据源的时候找不到想要的业务对象
[问题点数：40分，结帖人wl]
配置数据源的时候找不到想要的业务对象
[问题点数：40分，结帖人wl]
不显示删除回复
显示所有回复
显示星级回复
显示得分回复
只显示楼主
匿名用户不能发表回复！|An artificial neural network - back-propagation neural network &BP ANN&
Distant memory, hair up and share a nice ring, the format some chaos out. . .
A series of internal communication
An artificial neural network - back-propagation neural network (BP ANN)
* EDITORIAL words
A person's understanding will inevitably be biased, so the following remarks can not be all that you want to skeptical scrutiny. In the communication of ideas not scared of the purpose, the name may not be relevant to explain too much, are interested can check the information on in-depth look.
(Note: The artificial neural network is a big subject that branch of the following focused on BP ANN)
The computer player multiplayer, how to deal with and will make people feel like a real person in the Battle, and will not be too simple or too difficult to lose interest?
Write a state function to solve this problem:
Switch type of computer players
Case melee strength based
If (number of people friendly | | enemy attack power is weak | | full value of his life | | the enemy a short distance)
PK rushed to say
. . . . . .
Case type battle spell away
. . . . . .
. . . . . .
Indeed the state machine is used to solve the problem, but it looks like in order to do well that there are too many cases waiting for us. Moreover, the AI is dead, its behavior is born doomed, and should try to make a living!
If you have already thought of a way, first do not rush to look down, take the time to try to achieve it, do not let your PC world is too lonely.
2, or primer
(1) assumes known definition of the circle, which is more like the following round?
(Oval, round, square, three graphics)
(2) Suppose the square of the known 1999 = 3,996,001 3999 = 15,992,001 square
Then the square of 2999 =? (If we known the square = 0, 2,999 square =?)
We compared the time which is more rounded, graphics, and their awareness will be round to compare the rule of thumb or intuition, estimated that each map and circle the error, the smallest error is selected.
Mishap: If you choose the first two, but someone told you the first three is that they provided a perfect circle, then let you choose the next time, maybe you would choose the first three, because someone taught you that.
Guess when the square of 2999, based on previous learning experience and try, with an estimated value of around in the (3,996,001 + 15,992,001) / 2 is about 3000 square after that, speculation would be more accurate, if we know the square root of 2998 and may even be directly to guess right.
Mishap: But if only to tell you the square of 00,
to guess the square, the error will obviously be large, the probability of a direct guess is minimal.
The process described above may be biased, but in general BP neural network is similar to that from the point of view, to imitate the human brain processes, the first study sample set (1999 to 3999 square feet = 3,996,001 square while at the same time 3000 = 15,992,001 square = ), and constantly adjusting their weights to fit the sample set of results, attention is fitting that this small neural network, do not know what is a square (like we were young), tell it to give input A the results of B, he continued internal restructuring, to achieve the A was B, then told it was up to the C D, it will strive to fit, this time to ask what it is derived from the A? Could answer was not just B, so, the training continued until the error is less than the threshold, or the death of the cycle continues, unfortunately the (usually caused by poor training samples do not converge)
3, the simple definition
Said a lot is to be pointed out that the neural network flawed, so we're new to these ideas as their own when the same mistake why the problems can be designed to deal with an ANN, neural network, in fact, in accordance with a training sample, through a large number repeated calculations, build the solution and sample data from issues related to a formula.
Neural network a wonderful formula ≈
Story made rules, artificial neural network, information science and interdisciplinary branches of biology, the main idea is to simulate from a physiological point of view the human brain (nervous system), used in more extensive field of pattern recognition, such as handwriting recognition, weather forecasting , Identification of water quality, environmental quality evaluation.
There are a variety of neural network and is still ongoing rapid evolution, so to understand its main idea is the most critical. Widely used is also easier to understand the back-propagation neural network (Backpropagation ANN, referred to as BP).
Add that the traditional artificial intelligence, most of the people from the psychological point of simulation, such as chess and other human-computer game, through the heuristic algorithm, choose from the current situation can be translated into the best situation (or the local best situation), depth first, breadth-first algorithm, or by branch, can find a solution.
If using a neural network to solve this problem, do not expect it to be very strong, unless there are appropriate samples and methods, continuous training it. This is just as small to large, number of teachers under the guidance of our own process of gradual accumulation from ignorance.
4, a very important concept - the activation function (activation function)
The following quote from &AI Game Development&:
About 1011 adult brain neurons, and each neuron via synapses, receiving about 104 other neurons of the input potential. If all the input potential of the combined effect is strong enough, neurons would be activated, its activation potential pass it to other neurons. For a single neuron, its activation function to the total input in a nonlinear manner corresponding to the corresponding output value.
5, the ideal AI opponents or companions
Much ado, the code seen what happens now.
Neural networks composed by neurons, BP is commonly used in three-network, input layer, hidden layer, output layer, each neuron is constituted by the &input layer - hidden layer& and &hidden layer - output layer &between the neurons connected to each other, each connection has different weight (the weight training process which will be continuously adjusted), see the figure below.
(1) General level ---- NeuralNetworkLayer
BP network design, each layer can be regarded as a basic unit, a layer of abstraction common classes, properties include the following:
private int numberOfN / / The number of nodes neurons
private int numberOfChildN / / the lower the number of neuron nodes
private int numberOfParentN / / number of nodes in the upper neuron
private double [] [] / / Connect the two nodes on the lower array of rights re-
private double [] [] weightC / / the right to redo the adjustment of the correction array (to achieve the momentum method)
private double [] neuronV / / The layer of neurons calculated value (activation value)
private double [] desiredV / / layer neurons of the desired value (target)
private double [] / / the layer related to the error of each neuron
private double [] biasW / / the layer and each neuron weight-related bias
private double [] biasV / / the layer and each neuron related to the deviation (usually 1 or -1)
private double learningR / / learning rate used to calculate the weight correction
private boolean linearO / / if the layer neurons use linear activation function (output layer use)
private boolean useM / / adjust the weights whether to use momentum
private double momentumF / / momentum coefficient
private NeuralNetworkLayer parentL / / upper entity
private NeuralNetworkLayer childL / / lower layer entity
Not many important ways:
/ / Randomly initialize the weights (with the initial value to begin training)
public void randomizeWeights () {. . . . . . }
/ / Use input values of neurons and activation function, to calculate the layer of the value of each neuron activation or implicit value
public void calculateNeuronValues () {. . . . . . }
/ / Calculate the error of each neuron
public void calculateErrors () {. . . . . . }
/ / Adjust weights
public void adjustWeights () {. . . . . . }
(2) the overall network ---- NeuralNetwork
Properties are naturally included in the network layer:
private NeuralNetworkLayer inputL
private NeuralNetworkLayer hiddenL
private NeuralNetworkLayer outputL
/ / Set the input values used in the training set of input data and the practical application of the network input values set when the
public void setInput (int i, double value) {
/ / According to a set of input values, specify the output value of nature is spent training
public void setDesiredOutput (int i, double value) {
/ / According to a set of input values to produce output values
public void feedForward () {
/ / Calculate the error of a set of output values
public double calculateError () {
/ / Use back-propagation techniques, adjusting the connection weights
public void backPropagate () {
outputLayer.calculateErrors ();
hiddenLayer.calculateErrors ();
hiddenLayer.adjustWeights ();
inputLayer.adjustWeights ();
(3) network of trainers ---- Trainer
First need to build a reasonable training samples, this is crucial for a network, for example:
private double [] [] trainingSet = {
/ / Allowed to be hit by friendly enemy is fighting the number of clusters from the chase dodge
{0, 1, 0, 0.2, 0.9, 0.1, 0.1},
{1, 1, 1, 0.2, 0.9, 0.1, 0.1},
{0, 1, 0, 0.8, 0.9, 0.1, 0.1},
{0.1, 0.5, 0, 0.2, 0.9, 0.1, 0.1},
{0, 0.25, 1, 0.8, 0.1, 0.9, 0.1},
{0, 0.2, 1, 0.2, 0.1, 0.1, 0.9}
(Note that the value of each, were converted to [0,1] range, and very common practice)
With the training samples, the need to write a training method:
public void training () {
int c = 0;
double error = 1.0;
/ / Error is greater than the threshold or less than the threshold number of training and continue training
while ((error& 0.01) & & (c &50000)) {
error = 0;
for (i = 0; i &trainingSet. i + +) {
/ / Set the input
nn.setInput (0, trainingSet [i] [0]);
nn.setInput (1, trainingSet [i] [1]);
nn.setInput (2, trainingSet [i] [2]);
nn.setInput (3, trainingSet [i] [3]);
/ / Set the expected output
nn.setDesiredOutput (0, trainingSet [i] [4]);
nn.setDesiredOutput (1, trainingSet [i] [5]);
nn.setDesiredOutput (2, trainingSet [i] [6]);
/ / Generate output based on input
nn.feedForward ();
/ / Calculate the error of a set of output values, and add up
error + = nn.calculateError ();
/ / Use back-propagation techniques, adjusting the connection weights
nn.backPropagate ();
error = error / (float) / / error is the mean of the last to see
**** Training 14,784 times, the final error is 0.213308 ****
After training, you can save all the weight, so
also have training in it every time, because the weights are randomly initialized, specific training to find a weight, may be just a local optimum. Remember Mishap mentioned at the beginning ---- What tells you if only the square of 00,
to guess the square.
The method used, it is simple, set the input, calculate the output, get
public void using (float [] input) {
nn.setInput (0, input [0]);
nn.setInput (1, input [1]);
nn.setInput (2, input [2]);
nn.setInput (3, input [3]);
nn.feedForward ();
Trained relatively timid guy:
The number of enemy allies fighting the enemy from the value of life
Output layer, the final result, taking the maximum value for the final results:
Should be &enemy combatants,& the probability: = 0.345
Should be &closer to friendly forces,& the probability: = 0.3021347
Should be &quickly flash people,& the probability: = 0.67563
The number of enemy allies fighting the enemy from the value of life
Output layer, the final result, taking the maximum value for the final results:
Should be &enemy combatants,& the probability: = 0.98075
Should be &closer to friendly forces,& the probability: = 0.4573229
Should be &quickly flash people,& the probability: = 0.9983
The number of enemy allies fighting the enemy from the value of life
Output layer, the final result, taking the maximum value for the final results:
Should be &enemy combatants,& the probability: = 0.6574
Should be &closer to friendly forces,& the probability: = 0.503063
Should be &quickly flash people,& the probability: = 0.34931
The value of the number of lives fighting the enemy allies from the enemy
Output layer, the final result, taking the maximum value for the final results:
Should be &enemy combatants,& the probability: = 0.4205
Should be &closer to friendly forces,& the probability: = 0.60574
Should be &quickly flash people,& the probability: = 0.8959
These are a simple introductory example, the output reliability of the results may not be high, but I believe that when you successfully after it, there will be a joy, even shocking, in fact, many do still need to continue to learn to know also a lot.
Finally, do not mistakenly believe that neural network can only be a small toy, the network of neurons, the number of hidden layers can be more input values can be carbon dioxide content, particulate content, temperature and humidity, which tells us that air qua to the atmosphere various parameters, which tell us that it will rain tomorrow, the human brain is the highest agent, so the artificial intelligence and neural networks, worth the time to explore it a try.
6, Java class libraries:
Joone, has developed a long time, a graphical interface
7 Recommended Books
&Game Development Artificial Intelligence& &AI for Game Developers& (David M. Borug & Glenn Seemann)
&Game Programming in Artificial Intelligence Technology& &AI Techniques for Game Programming& (Mat Buckland)
8, the algorithm steps
1, with a small random value will be the right initialization
2, select an input mode, add the input layer
3, the signal is transmitted through the network before
4, the actual output and expected output compare
5, the direction of propagation through the error, the error before the calculation of a layer of
6, updating the weights of all connections
7, for another input, return to the second step and repeat the process
(That is updated using the previous weight, so after a preference for the input of the situation, grass is always greener)
9, advantages and disadvantages:
1, the advantages:
(1) a 3-layer BP network can be done to arbitrary n-dimensional mapping of m-dimensional
2, shortcomings
(1) nonlinear optimization, the situation encountered local minimum
(2) slow convergence, more than a thousand steps
(3) information forward and backward flow through the network, and use of derivatives, the biological point of view the lack of credibility
(4) the number of network hidden nodes selected by experience
(5) The grass is always greener, and the information contained in each sample must be the same number of
1, to prevent over-fitting
(1) minimize the number of neurons
(2) by adding swing (jitter)
(3) early termination
Please enable JavaScript to view the
The recent project needs, need to use socket, demand is such, there are many sets of data collection server, and another three used to monitor the server, monitoring server, message server to collect and monitor the server receives the message, chang
Content Outline 1 What is the reconstruction of (what) (2) the timing of the reconstruction (when) 3 How to reconstruct (how) 4. Some important commonly used reconstruction method 5. Refactoring to introduce a model by refactoring Download
Some companies test pen Sony T test 1. Complete the following procedures * *.*. *..*..*.. *...*...*...*... *....*....*....*....*.... *.....*.....*.....*.....*.....*..... *......*......*......*......*......*......*...... *.......*.......*.......*.....
* DC: a branch of biological neurons, the cells pass information between different parts, dendritic cells normally responsible for the input function, although there are many dendritic output function of neurons in the ANN input analog DC * Determini
DOS command / port (A) MD - creates a subdirectory 1. Function: create a new subdirectory 2. Type: Internal command 3. Format: MD [drive:] [pathname] &directory name& 4. Instructions for use: (1) &letter&: Specifies the subdirectory to cre
We all know each Hadoop node in the internal communication it is necessary to ensure that the two versions is the same (0.19 is the case, more than 0.20 do not know). That JobTracker and TaskTracker must use the same compiler version, and if not, may
The following look BaseForm using S using System.Collections.G using System.ComponentM using System.D using System.D using System.L using System.T using System.Windows.F using yokohama. using System.Timer
In recent years, the game industry led to the rapid development of the game AI (Artificial Intelligence, referred to as AI) development, more and more of the game using artificial intelligence technology to improve gameplay. In the electronic game, p
http://blog.csdn.net/ityuany/archive//5509750.aspx In recent years, the game industry's rapid development led to game artificial intelligence (Artificial Intelligence, referred to as AI) development, more and more of the game using artifici
Copyright (C) , All Rights Reserved.
版权所有 闽ICP备号
processed in 0.042 (s). 14 q(s)