Tafiti Search Visualization Deployment Guide

本文档提供了详细的Tafiti搜索可视化应用部署步骤,包括Windows Live账户设置、数据库配置、Web应用程序设置等内容。介绍了如何使用SQLCMD或SQL Server Management Studio进行数据库设置,并配置了LiveID Web认证、数据库连接字符串等关键参数。

摘要生成于 C知道 ,由 DeepSeek-R1 满血版支持, 前往体验 >

Tafiti Search Visualization Deployment Guide 

导读:
  This deployment guide is very large (~ 3-4 MB). You can download an XPS of this Wiki Page from here(NOTE: this file may get out of sync with the Wiki).
  Configuration
  Windows Live Account Setup
  Database setup 每 without Management Tools (SQLCMD)
  Database setup 每 with Management Tools
  Web Application Setup
  Solution Overview (Projects)
  Configuring Live ID Web Authentication
  Configuring Database Connection String
  Configuring Search API settings
  Configuring AdCenter Analytics
  Building the project
  Executing &Smoke testing
  Start the project
  Testing Shelves
  Deployment options
  Demonstration Script
  Demonstration machine configuration
  Apple Mac
  PC (Windows Vista/Windows XP)
  Demonstration Walkthrough
  FAQ / More information
  Server / Development Environment Setup / Pre-requisites Install the following on your computer (Windows Server 2003, Windows XP, and Windows Vista or later):
  IIS 6.0 or later
  If IIS 6.0, you'll need to configure it for Silverlight (see http://msdn2.microsoft.com/en-us/library/ms752346.aspx)
  SQL Server 2005 (http://www.microsoft.com/sql/downloads/trial-software.mspx)
  NOTE: You can download SQL Express for free from hereand the SQL Express Management tools from here
  .NET Framework 2.0 (Windows Update)
  .NET Framework 3.0 (Windows Update)
  ASP.NET 2.0 AJAX Extensions
  ASP.NET AJAX Control Toolkit
  Microsoft AntiXSS Library
  Microsoft Visual Studio 2008or Visual Web Developer 2008 (free)
  Configuration
  Windows Live Account SetupYou will need to gain access and/or credentials from each of the following, which will be stored in the Web.config of the Web project:
  Windows Live Authentication (http://msm.live.com/app) (see the web.config for redirect URLs - http://localhost:50008/WLQuickApps.Tafiti.WebSite/Register.aspx if you are running in local debug mode)
  Time Lapse in setting up account: Instant
  
  
  Database setup 每 without Management Tools (SQLCMD)You can configure the database via the Command Line if you do not have the SQL Server Management Studio installed. This can be done via the SQLCMD command line (tutorial).
  Open a command prompt and change the directory to:
  C:/Program Files/Microsoft SQL Server/90/Tools/Binn
  To connect with the SQL instance type:
  sqlcmd 每S ./SQLExpress
  
  
  Paste the following statement and press ENTER. NOTE: These databases will be created in C:/temp if you wish for them to be created elsewhere modify the script before you paste it into the command window.CREATE DATABASE TafitiON
  PRIMARY ( NAME = N'Tafiti', FILENAME = N'c:/temp/Tafiti.mdf' , SIZE = 41984KB , MAXSIZE = UNLIMITED, FILEGROWTH = 1024KB )
  LOG ON ( NAME = N'Tafiti_log', FILENAME = N'c:/temp/Tafiti_log.ldf' , SIZE = 22144KB , MAXSIZE = 2048GB , FILEGROWTH = 10%)
  GO
  
  
  Quit the existing SQLCMD session by typing Quitand press ENTER
  
  
  Create the objects using the Create.sql by using the 每i (InputFile) parameter
  sqlcmd 每s ./SQLExpress 每i C:/Code/Codeplex/WLQuickApps/WLQuickApps.Tafiti/create.sql
  
  
  
  
  Database setup 每 with Management Tools (SQL Server Management Studio)If you have the SQL Server Management Studio installed you can create a new database and configure the tables.
  Open the SQL Server Management Studio
  
  
  Create a database called Tafiti
  
  
  
  
  Run the create.sqlscript found in the project root to set up the tables NOTE: be sure to run this against the Tafiti database and not the master database.
  Open the create.sqlfile in the SQL Management studio (File --> Open --> File) or CTRL+O
  
  
  
  
  
  
  Click Execute to run the script
  
  
  See the command has completed successfully
  Web Application SetupThe web application has several settings in the web.configwhich must be configured to make this application successfully execute.
  Open your development Environment (Visual Web Developer or Visual Studio 2008)
  
  
  Open the solution file
  
  
  Solution Overview (Projects)
  
  
  The solution will open with 3 projects:
  Website 每 the presentation layer
  Business 每 the business layer which stores non presentation specific logic
  Data 每 the data layer manages the persistence of objects back into the data provider (in this case SQL Server)
  Configuring Live ID Web AuthenticationBe sure to update Web.configwith the credentials obtained from the Windows Live? ID configuration steps above. You must copy and paste the Windows Live App ID and the Secret Key
  
  
  
  
  Configuring Database Connection String
  Verify the connection string is correct by default it uses ./SQLExpress with Integrated security.
  You must also copy the connection string into   
  
  Configuring Search API settings
  Generate a Search API developer key at http://search.msn.com/developer
  
  
  
  
  
  
  Embed the search API key in the Web.Config
  
  
  Configuring AdCenter AnalyticsIf you have an AdCenter analytics account or would like your usage of Tafiti to be reported set this field. If you don*t want to report usage of the Tafiti code set the field as blank. AdCenter Analytics accounts are not broadly available yet.
  
  
  Building the project
  
  
  Executing &Smoke testingTo test Tafiti Search Visualization you need to run the code and step over the key areas such as:
  UX Rendering 每 does the website render
  Search API Queries 每 is the search API accessible and has the API Key been successfully set
  Authentication 每 has the Windows Live? ID Web Authentication been configured correctly
  Storage 每 has the SQL database connection string and objects been created/configured correctly.
  Start the projectYou can either F5 to start debugging or right click on the default.aspx to start the local web server.
  
  
  
  
  NOTE: It may take a couple of minutes for the Search API key to become active, during this time you will see: An Error has occurred.
  
  
  Testing ShelvesA shelf is a saved set of search results. You can test these by authenticating and dragging search results to the right hand site. This will test the following piece of technology: Live ID authentication, SQL connection &SQL schema.
  
  
  
  
  NOTE: If you see this error after authentication means you are redirecting to a different port than the debugger is running on. Change this by selecting the Website project and then the property window and change the dynamic ports to FALSE and 50008
  
  
  
  
  
  
  Sign in and drag a result onto the right hand side
  
  
  Close the browser and reopen and sign in and the shelf should be fully populated.
  
  
  The shelf will be created in the database.
  
  
  Deployment options
  Copy/publish the output of the Web project to the virtual directory set up earlier for Web hosting
  Run the site in local debug mode (the Web Site has been set to non dynamic ports)

本文转自
http://www.codeplex.com/WLQuickApps/Wiki/View.aspx?title=Tafiti%20Search%20Visualization&referringTitle=Home

**LSTM时间序列神经网络预测MATLAB代码** 在人工智能领域,长短期记忆网络(Long Short-Term Memory,简称LSTM)是一种特殊的循环神经网络(Recurrent Neural Network, RNN),特别适合处理时间序列数据,如股票价格预测、天气预报、语音识别等。MATLAB作为强大的科学计算工具,提供了丰富的深度学习库,包括对LSTM模型的支持,使得在MATLAB环境中构建和训练LSTM模型成为可能。 我们需要理解LSTM的基本结构。LSTM由单元细胞、输入门、输出门和遗忘门组成,这些门控机制能够有效地解决传统RNN的梯度消失问题,从而更好地捕捉长期依赖性。在MATLAB中,我们可以通过`nnlstm`函数来创建LSTM层,并用`trainNetwork`来训练模型。 描述中提到“可以直接运行”,这意味着提供的MATLAB代码应该包含完整的数据预处理、模型构建、训练和预测过程。数据预处理是关键步骤,包括数据清洗、归一化、切分训练集和测试集等。在处理时间序列数据时,通常会使用滑动窗口技术将连续的数据点转化为多个固定长度的序列样本。 模型构建阶段,我们需要定义LSTM网络的架构,包括输入层、LSTM层、全连接层和输出层。MATLAB的`nnlstm`函数允许我们设置单元数量、隐藏层层数以及是否启用双向LSTM。同时,我们还需要选择合适的损失函数(如均方误差)和优化器(如Adam或随机梯度下降)。 训练过程使用`trainNetwork`函数,可以指定训练迭代次数、批次大小、学习率等参数。训练过程中可能需要监控损失函数的变化和验证集上的性能,以便调整超参数和防止过拟合。 预测阶段,我们会用训练好的LSTM模型对未来的时间序列进行预测。MATLAB的`predict`函数可以生成预测值,然后可能需要对这些预测结果进行反归一化,以便与原始数据尺度匹配。 在实际应用中,LSTM模型的性能往往取决于数据的质量和特征工程,以及模型的结构和训练策略。因此,对于给定的MATLAB代码,我们应当关注以下几点: 1. 数据预处理的细节:如何处理缺失值,如何进行归一化,以及如何构建时间序列样本。 2. LSTM网络的架构:单元数量、层数、是否双向,以及全连接层的配置。 3. 训练过程的设置:优化器的选择,学习率调度,以及早停策略。 4. 预测和后处理:如何生成预测序列,以及如何将预测结果还原到原始尺度。 由于没有具体的代码内容,以上分析基于一般LSTM模型的构建和训练流程。实际的MATLAB代码会包含这些具体实现,通过阅读和理解代码,我们可以深入学习LSTM在时间序列预测中的应用,以及如何在MATLAB环境中高效地实现这一过程。
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值