CS代写归档 - avatest

2023年5月25日2023年5月25日 CS代写, Machine Learning, 机器学习, 计算机代写

计算机代写|机器学习代写Machine Learning代考|Don’t waste our time: Meeting with cross-functional teams

如果你也在怎样代写机器学习Machine Learning 这个学科遇到相关的难题，请随时右上角联系我们的24/7代写客服。机器学习Machine Learning是一个致力于理解和建立 “学习 “方法的研究领域，也就是说，利用数据来提高某些任务的性能的方法。机器学习算法基于样本数据（称为训练数据）建立模型，以便在没有明确编程的情况下做出预测或决定。机器学习算法被广泛用于各种应用，如医学、电子邮件过滤、语音识别和计算机视觉，在这些应用中，开发传统算法来执行所需任务是困难的或不可行的。

机器学习Machine Learning程序可以在没有明确编程的情况下执行任务。它涉及到计算机从提供的数据中学习，从而执行某些任务。对于分配给计算机的简单任务，有可能通过编程算法告诉机器如何执行解决手头问题所需的所有步骤；就计算机而言，不需要学习。对于更高级的任务，由人类手动创建所需的算法可能是一个挑战。在实践中，帮助机器开发自己的算法，而不是让人类程序员指定每一个需要的步骤，可能会变得更加有效。

机器学习Machine Learning代写，免费提交作业要求，满意后付款，成绩80\%以下全额退款，安全省心无顾虑。专业硕博写手团队，所有订单可靠准时，保证 100% 原创。最高质量的机器学习Machine Learning作业代写，服务覆盖北美、欧洲、澳洲等国家。在代写价格方面，考虑到同学们的经济条件，在保障代写质量的前提下，我们为客户提供最合理的价格。由于作业种类很多，同时其中的大部分作业在字数上都没有具体要求，因此机器学习Machine Learning作业代写的价格不固定。通常在专家查看完作业要求之后会给出报价。作业难度和截止日期对价格也有很大的影响。

avatest™帮您通过考试

avatest™的各个学科专家已帮了学生顺利通过达上千场考试。我们保证您快速准时完成各时长和类型的考试，包括in class、take home、online、proctor。写手整理各样的资源来或按照您学校的资料教您，创造模拟试题，提供所有的问题例子，以保证您在真实考试中取得的通过率是85%以上。如果您有即将到来的每周、季考、期中或期末考试，我们都能帮助您！

在不断发展的过程中，avatest™如今已经成长为论文代写，留学生作业代写服务行业的翘楚和国际领先的教育集团。全体成员以诚信为圆心，以专业为半径，以贴心的服务时刻陪伴着您，用专业的力量帮助国外学子取得学业上的成功。

•最快12小时交付

•200+ 英语母语导师

•70分以下全额退款

想知道您作业确定的价格吗? 免费下单以相关学科的专家能了解具体的要求之后在1-3个小时就提出价格。专家的报价比上列的价格能便宜好几倍。

我们在计算机Quantum computer代写方面已经树立了自己的口碑, 保证靠谱, 高质且原创的计算机Quantum computer代写服务。我们的专家在机器学习Machine Learning代写方面经验极为丰富，各种机器学习Machine Learning相关的作业也就用不着说。

计算机代写|机器学习代写Machine Learning代考|Don’t waste our time: Meeting with cross-functional teams

Chapter 3, in discussing the planning and experimentation phases of a project, noted that one of the most important aspects to keep in mind (aside from the ML work itself) is the communication during those phases. The feedback and evaluation received can be an invaluable tool to ensure that the MVP gets delivered on time and is as correct as can be so that the full development effort can proceed.

Let’s take another look at our Gantt chart from figure 4.7 for keeping track of the highlevel progress of each team’s work throughout the phases. For the purposes of communication, however, we’re concerned with only the top portion, shown in figure 4.9 .

Depending on the type of project being built, countless more meetings may be spread throughout the phases (as well as follow-up meetings for months after release to review metrics, statistics, and estimations of the resiliency of the solution). Even if the development phase takes nine months, for instance, the biweekly progress report meetings are just repetitive discussions on the progress of accomplishments during the previous sprint. We’re going to break down these phases in detail next.

计算机代写|机器学习代写Machine Learning代考|Experimental update meeting: Do we know what we’re doing here?

The experimental update meeting is the one that the DS team dreads more than any other, and the meeting that everyone else is incredibly excited for. The meeting interrupts the DS team in the midst of half-baked prototype implementations and unfinished research. The state of elements in flux is at nearly peak entropy.

This meeting is perhaps the second most important meeting in a project, though. This is the second-to-last time for the team members to have the ability to graciously raise a white flag in surrender if they’ve discovered that the project is untenable, will end up taking more time and money than the team has allocated, or is of such complexity that technologies will not be invented within the next 50 years to meet the requirements set forth. This is a time for honesty and reflection. It’s a time to set one’s ego aside and admit defeat, should the situation call for it.

The overriding question dominating this discussion should be, “Can we actually figure this out?” Any other discussions or ideations about the project are completely irrelevant at this point. It is up to the DS team to report on the status of its discoveries (without getting into the weeds of model-specific details or additional algorithms that they will be employing for testing, for example). The most critical discussion points for this meeting should be the following:
” How is the progress toward the prototype coming along?

Have you figured out any of the things that you’re testing yet?
Which one looks like it’s the most promising so far?
Are you going to stop pursuing anything that you had planned to test?
Are we on track to have a prototype by the scheduled due date?
What risks have you uncovered so far?
Are there challenges with the data that the DE team needs to be made aware of?
Are we going to need a new technology, platform, or tooling that the team isn’t familiar with?
As of right now, do you feel as though this is a solvable problem for us?

机器学习代写

计算机代写|机器学习代写Machine Learning代考|Don’t waste our time: Meeting with cross-functional teams

第3章，在讨论项目的计划和实验阶段时，注意到要记住的最重要的方面之一(除了ML工作本身)是这些阶段的沟通。收到的反馈和评估是一个非常宝贵的工具，可以确保MVP按时交付，并且尽可能正确，以便完整的开发工作可以继续进行。

让我们再看一下图4.7中的甘特图，以跟踪每个团队在整个阶段的工作的高级进展。但是，出于通信的目的，我们只关注上面的部分，如图4.9所示。

根据正在构建的项目的类型，无数的会议可能会分布在各个阶段(以及发布后几个月的后续会议，以审查解决方案的度量、统计数据和评估)。例如，即使开发阶段需要9个月的时间，两周一次的进度报告会议也只是重复讨论前一个冲刺期间的成就进展。接下来我们将详细分解这些阶段。

计算机代写|机器学习代写Machine Learning代考|Experimental update meeting: Do we know what we’re doing here?

实验更新会议是DS团队最害怕的会议，也是每个人都无比兴奋的会议。会议打断了DS团队半生不熟的原型实现和未完成的研究。流动中的元素的状态几乎处于熵的峰值。

然而，这个会议可能是项目中第二重要的会议。如果团队成员发现项目是站不住脚的，最终将花费比团队分配的更多的时间和金钱，或者是如此复杂以至于在未来50年内无法发明技术来满足所设定的需求，那么这是团队成员有能力优雅地举起白旗投降的倒数第二次。这是一个诚实和反思的时刻。这是一个把自我放在一边，承认失败的时候，如果情况需要的话。

主导这个讨论的最重要的问题应该是，“我们真的能解决这个问题吗?”在这一点上，关于项目的任何其他讨论或想法都是完全无关的。这取决于DS团队报告其发现的状态(例如，不涉及特定于模型的细节或他们将用于测试的额外算法)。本次会议最关键的讨论要点应如下:
“原型机的进展如何?”

你想好要测试的东西了吗?

到目前为止，哪一个看起来最有希望?

你打算停止任何你计划测试的东西吗?

我们能在预定的截止日期前做出样品吗?

到目前为止，你发现了哪些风险?

数据方面是否存在需要让DE团队了解的挑战?

我们是否需要团队不熟悉的新技术、平台或工具?

到目前为止，你觉得这对我们来说是一个可以解决的问题吗?

计算机代写|机器学习代写Machine Learning代考请认准UprivateTA™. UprivateTA™为您的留学生涯保驾护航。

微观经济学代写

微观经济学是主流经济学的一个分支，研究个人和企业在做出有关稀缺资源分配的决策时的行为以及这些个人和企业之间的相互作用。my-assignmentexpert™ 为您的留学生涯保驾护航在数学Mathematics作业代写方面已经树立了自己的口碑, 保证靠谱, 高质且原创的数学Mathematics代写服务。我们的专家在图论代写Graph Theory代写方面经验极为丰富，各种图论代写Graph Theory相关的作业也就用不着说。

线性代数代写

线性代数是数学的一个分支，涉及线性方程，如：线性图，如：以及它们在向量空间和通过矩阵的表示。线性代数是几乎所有数学领域的核心。

博弈论代写

现代博弈论始于约翰-冯-诺伊曼（John von Neumann）提出的两人零和博弈中的混合策略均衡的观点及其证明。冯-诺依曼的原始证明使用了关于连续映射到紧凑凸集的布劳威尔定点定理，这成为博弈论和数学经济学的标准方法。在他的论文之后，1944年，他与奥斯卡-莫根斯特恩（Oskar Morgenstern）共同撰写了《游戏和经济行为理论》一书，该书考虑了几个参与者的合作游戏。这本书的第二版提供了预期效用的公理理论，使数理统计学家和经济学家能够处理不确定性下的决策。

微积分代写

微积分，最初被称为无穷小微积分或 “无穷小的微积分”，是对连续变化的数学研究，就像几何学是对形状的研究，而代数是对算术运算的概括研究一样。

它有两个主要分支，微分和积分；微分涉及瞬时变化率和曲线的斜率，而积分涉及数量的累积，以及曲线下或曲线之间的面积。这两个分支通过微积分的基本定理相互联系，它们利用了无限序列和无限级数收敛到一个明确定义的极限的基本概念。

计量经济学代写

什么是计量经济学？
计量经济学是统计学和数学模型的定量应用，使用数据来发展理论或测试经济学中的现有假设，并根据历史数据预测未来趋势。它对现实世界的数据进行统计试验，然后将结果与被测试的理论进行比较和对比。

根据你是对测试现有理论感兴趣，还是对利用现有数据在这些观察的基础上提出新的假设感兴趣，计量经济学可以细分为两大类：理论和应用。那些经常从事这种实践的人通常被称为计量经济学家。

MATLAB代写

MATLAB 是一种用于技术计算的高性能语言。它将计算、可视化和编程集成在一个易于使用的环境中，其中问题和解决方案以熟悉的数学符号表示。典型用途包括：数学和计算算法开发建模、仿真和原型制作数据分析、探索和可视化科学和工程图形应用程序开发，包括图形用户界面构建MATLAB 是一个交互式系统，其基本数据元素是一个不需要维度的数组。这使您可以解决许多技术计算问题，尤其是那些具有矩阵和向量公式的问题，而只需用 C 或 Fortran 等标量非交互式语言编写程序所需的时间的一小部分。MATLAB 名称代表矩阵实验室。MATLAB 最初的编写目的是提供对由 LINPACK 和 EISPACK 项目开发的矩阵软件的轻松访问，这两个项目共同代表了矩阵计算软件的最新技术。MATLAB 经过多年的发展，得到了许多用户的投入。在大学环境中，它是数学、工程和科学入门和高级课程的标准教学工具。在工业领域，MATLAB 是高效研究、开发和分析的首选工具。MATLAB 具有一系列称为工具箱的特定于应用程序的解决方案。对于大多数 MATLAB 用户来说非常重要，工具箱允许您学习和应用专业技术。工具箱是 MATLAB 函数（M 文件）的综合集合，可扩展 MATLAB 环境以解决特定类别的问题。可用工具箱的领域包括信号处理、控制系统、神经网络、模糊逻辑、小波、仿真等。

2023年5月25日2023年5月25日 CS代写, Machine Learning, 机器学习, 计算机代写

计算机代写|机器学习代写Machine Learning代考|POST-RESEARCH PHASE DISCUSSION (UPDATE MEETING)

avatest™帮您通过考试

•最快12小时交付

•200+ 英语母语导师

•70分以下全额退款

想知道您作业确定的价格吗? 免费下单以相关学科的专家能了解具体的要求之后在1-3个小时就提出价格。专家的报价比上列的价格能便宜好几倍。

计算机代写|机器学习代写Machine Learning代考|POST-RESEARCH PHASE DISCUSSION (UPDATE MEETING)

For the sake of example within our scenario, let’s assume that the DS team identifies that two models must be built to satisfy the requirements from the planning phase user-journey simulation. Based on the team members’ research, they decide that they want to pit both collaborative filtering and frequent-pattern-growth (FP-growth) marketbasket analysis algorithms against deep learning implementations to see which provides a higher accuracy and lower cost of ownership for retraining.

The DS lead assigns two groups of data scientists and ML engineers to work on these competing implementations. Both groups generate simulations of the model results on the exact same synthetic customer dataset, providing mock product images to a wireframe of the pages displaying these recommendations for the actual website.
This meeting should not focus on any of the implementation details. Instead, it should focus solely on the results of the research phase: the whittling down of nighinfinite options that have been read about, studied, and played with. The team has found a lot of great ideas and an even larger group of potential solutions that won’t work based on the data available, and has reduced the list of great ideas to a bake-off of two implementations that they’ll pit against each other. Don’t bring up all of the options that you’ve explored. Don’t mention something that has amazing results but will likely take two years to build. Instead, distill the discussion to the core details required to get the next phase going: experimentation.

Show these two options to the SMEs, solely within the confines of presenting what can be done with each algorithmic solution, what is impossible with one or both, and when the SMEs can expect to see a prototype in order to decide which they like better. If no discernable difference exists in the quality of the predictions, the decision of which to go with should be based on the drawbacks of the approaches, leaving the technical complexity or implementation details out of the discussion.

Keep the discussion in these dense meetings focused on relatable language and references that your audience will comprehend and associate with. You can do the translating in your head and leave it there. The technical details should be discussed only internally by the DS team, the architect, and engineering management.

In many cases that I’ve been involved with, the experimental testing phase may test out a dozen ideas but present only the two most acceptable to a business unit for review. If the implementation would be overly onerous, costly, or complex, it’s best to present options that will guarantee the greatest chance of project success-even if they’re not as fancy or exciting as other solutions. Remember: the DS team has to maintain the solution, and something that sounds really cool during experimentation can turn into a nightmare to maintain.

计算机代写|机器学习代写Machine Learning代考|POST-EXPERIMENTATION PHASE (SME/UAT REVIEW)

Following the experimentation phase, the subteams within the DS group build two prototypes for the recommendation engine. In the previous milestone meeting, the options for both were discussed, with their weaknesses and strengths presented in a way that the audience could understand. Now it’s time to lay the prediction cards out on the table and show off what a prototype of the solution looks like.

Before, during reviews of the potential solutions, some pretty rough predictions were shown. Duplicate products with different product IDs were right next to one another, endless lists of one product type were generated for some users (there’s no way that anyone likes belts that much), and the list of critical issues with the demo were listed out for consideration. In those first early pre-prototypes, the business logic and feature requirements weren’t built out yet, since those elements directly depended on the models’ platform and technology selection.

The goal of the presentation that completes the experimentation phase should be to show a mock-up of the core features. Perhaps elements need to be ordered based on relevancy. Special considerations may require recommending items based on price point, recent non-session-based historical browsing, and the theory that certain customers have implicit loyalty to certain brands. Each of these agreed-upon features should be shown to the entire team. The full implementation, however, should not be done by this point, but merely simulated to show what the eventual designed system would look like.
The results of this meeting should be similar to those from the initial planning meeting: additional features that weren’t recognized as important can be added to the development planning, and if any of the original features are found to be unnecessary, they should be removed from the plan. Revisiting the original plan, an updated user experience might look something like figure 4.8 .

机器学习代写

计算机代写|机器学习代写Machine Learning代考|POST-RESEARCH PHASE DISCUSSION (UPDATE MEETING)

为了在我们的场景中提供示例，让我们假设DS团队确定必须构建两个模型来满足规划阶段用户旅程模拟的需求。根据团队成员的研究，他们决定将协同过滤和频繁模式增长(FP-growth)市场篮子分析算法与深度学习实现进行比较，看看哪种算法为再培训提供了更高的准确性和更低的拥有成本。

DS的领导分配了两组数据科学家和ML工程师来研究这些相互竞争的实现。两个小组都在完全相同的合成客户数据集上生成模型结果的模拟，为实际网站显示这些推荐的页面的线框提供模拟产品图像。
这次会议不应集中讨论任何实施细节。相反，它应该专注于研究阶段的结果:减少那些已经阅读、研究和体验过的近乎无限的选择。团队已经发现了许多很棒的想法和更大的潜在解决方案，但这些想法并不基于现有的数据，并且已经将这些想法的列表减少到两个实现的比拼，他们将相互竞争。不要把你探索过的所有选项都提出来。不要提及那些有惊人结果但可能需要两年时间才能完成的事情。相反，我们应该将讨论浓缩到进入下一阶段(实验)所需的核心细节上。

向中小企业展示这两种选择，仅在展示每种算法解决方案可以完成的范围内，其中一种或两种算法解决方案不可能完成的范围内，以及中小企业何时可以期望看到原型，以便决定他们更喜欢哪一种。如果预测的质量没有明显的差异，那么应该根据方法的缺点来决定采用哪一种方法，而不讨论技术复杂性或实现细节。

在这些密集的会议中，让讨论集中在听众能够理解和联想到的相关语言和参考资料上。你可以在脑子里做翻译，把它留在那里。技术细节应该只在内部由DS团队、架构师和工程管理人员讨论。

在我参与的许多案例中，实验测试阶段可能会测试一打想法，但只向业务部门提供最可接受的两个想法进行审查。如果实现过于繁重、昂贵或复杂，那么最好提供能够保证项目成功的最大机会的选项——即使它们不像其他解决方案那样花哨或令人兴奋。请记住:DS团队必须维护解决方案，在实验过程中听起来很酷的东西可能会变成维护的噩梦。

计算机代写|机器学习代写Machine Learning代考|POST-EXPERIMENTATION PHASE (SME/UAT REVIEW)

在实验阶段之后，DS组中的子团队为推荐引擎构建了两个原型。在之前的里程碑会议上，讨论了这两种选择，并以听众能够理解的方式展示了它们的优缺点。现在是时候把预测卡片放在桌子上，展示解决方案的原型是什么样子的了。

之前，在对潜在解决方案的审查期间，给出了一些相当粗略的预测。具有不同产品id的重复产品彼此相邻，为某些用户生成了一种产品类型的无穷无尽的列表(没有人会那么喜欢皮带)，并且列出了演示的关键问题列表以供考虑。在最初的预原型中，业务逻辑和功能需求还没有构建出来，因为这些元素直接依赖于模型的平台和技术选择。

完成实验阶段的演示的目标应该是展示核心功能的模型。也许需要根据相关性对元素进行排序。特殊的考虑可能需要基于价格点、最近的非会话历史浏览以及某些客户对某些品牌有隐性忠诚度的理论来推荐商品。每个商定的特性都应该展示给整个团队。然而，完整的实现不应该在此时完成，而只是模拟以显示最终设计的系统的样子。
这次会议的结果应该与最初的规划会议的结果相似:可以将不重要的附加功能添加到开发规划中，如果发现任何原始功能是不必要的，则应将其从计划中删除。重新查看原始计划，更新后的用户体验可能如图4.8所示。

计算机代写|机器学习代写Machine Learning代考请认准UprivateTA™. UprivateTA™为您的留学生涯保驾护航。

微观经济学代写

线性代数代写

线性代数是数学的一个分支，涉及线性方程，如：线性图，如：以及它们在向量空间和通过矩阵的表示。线性代数是几乎所有数学领域的核心。

博弈论代写

微积分代写

计量经济学代写

MATLAB代写

2023年5月25日2023年5月25日 CS代写, Machine Learning, 机器学习, 计算机代写

计算机代写|机器学习代写Machine Learning代考|WHAT DOES THE IDEAL END-STATE LOOK LIKE?

avatest™帮您通过考试

•最快12小时交付

•200+ 英语母语导师

•70分以下全额退款

想知道您作业确定的价格吗? 免费下单以相关学科的专家能了解具体的要求之后在1-3个小时就提出价格。专家的报价比上列的价格能便宜好几倍。

计算机代写|机器学习代写Machine Learning代考|WHAT DOES THE IDEAL END-STATE LOOK LIKE?

The ideal implementation is hard to define at first (particularly before any experimentation is done), but it’s incredibly useful to the experimentation team to hear all aspects of an ideal state. During these open-ended stream-of-consciousness discussions, a tendency of most ML practitioners is to instantly decide what is and isn’t possible based on the ideas of people who don’t understand what ML is. My advice is to simply listen. Instead of shutting down a thread of conversation immediately as being out of scope or impossible, let the conversation happen.

You may find an alternative path during this creative ideation session that you otherwise would have missed. You might just find a simpler, less unique, and far more maintainable ML solution than what you may have come up with on your own. The most successful projects that I’ve worked on over the years have come from having these sorts of creative discussions with a broad team of SMEs (and, when I’ve been lucky, the actual end users) to allow me to shift my thinking into creative ways of getting as close as possible to their vision.

Discussing an ideal end state isn’t just for the benefit of a more amazing ML solution, though. Engaging the person asking for the project to be built allows their perspective, ideas, and creativity to influence the project in positive ways. The discussion also helps build trust and a feeling of ownership in the development of the project that can help bring a team together.

Learning to listen closely to the needs of your ML project’s customer is one of the most important skills of an ML engineer-far more than mastering any algorithm, language, or platform. It will help guide what you’re going to try, what you’re going to research, and how to think differently about problems to come up with the best solution that you possibly can.

In the scenario shown in figure 4.4 , the initial planning meeting results in a rough sketch of the ideal state. This likely will not be the final engine (based on my experience, that most certainly is never the case). But this diagram will inform how to convert those functional blocks into systems. It will help inform the direction of experimentation, as well as the areas of the project that you and the team will need to research thoroughly to minimize or prevent unexpected scope creep, as shown in figure 4.5 .

计算机代写|机器学习代写Machine Learning代考|Scope creep: An almost guaranteed assassination of a project

Improper planning (or planning without involving the team that the project is being built for) is a perfect recipe for one of the most frustrating ways of having a project die a slow, whimpering death. Also known as ML death by a thousand requests, this concept materializes at later stages of development, particularly when a demo is shown to a team that is uninformed about the details that went into building the project. If the customer (the internal business unit) was not party to the planning discussions, they inevitably will have questions, and many of them, about what the demo does.
In nearly every case that l’ve seen (or caused in my earlier days of trying to “hero my way through” a project without asking for input), the result of the demo session is going to be dozens of requests for additional features and requirements to be added. This is expected (even in a properly designed and planned project), but if the implementation is unable to easily include critical features that relate to immutable business operation “laws,” a potential full reimplementation of the project could be required. That leaves decision makers with the difficult choice of whether to delay the project because of the decision that the DS team (or individual) made, or to scrap the project entirely to prevent the chances of a repeat of the initial failure.
Few things are more devastating to hear in the world of ML than intensely negative feedback immediately after something goes live in production. Getting a flood of email from executive-level staff indicating that the solution you just shipped is recommending cat toys to a dog owner is laughable, but one that is recommending adultthemed products to children is about as bad as it can get. The only thing worse is realizing, right before the project is shipped, during user-acceptance testing (UAT), that an insurmountable list of changes needs to be made to satisfy the urgent requirements of the business and that it would take less time to start the project over from scratch than to make the changes to the existing solution.
Identifying scope creep is important, but its magnitude can be minimized, and in some cases eliminated. The appropriate level of discussion needs to be reached, and critical aspects of a project included in sometimes excruciating recursive and painful detail well before a single character is typed in an experimentation notebook or IDE.

机器学习代写

计算机代写|机器学习代写Machine Learning代考|WHAT DOES THE IDEAL END-STATE LOOK LIKE?

理想的实现在一开始很难定义(特别是在任何实验完成之前)，但是对于实验团队来说，听到理想状态的所有方面是非常有用的。在这些开放式的意识流讨论中，大多数ML实践者倾向于根据不了解ML是什么的人的想法立即决定什么是可能的，什么是不可能的。我的建议就是倾听。不要因为超出范围或不可能而立即停止对话，让对话开始吧。

你可能会在这个创造性的构思过程中找到一条你可能会错过的替代路径。您可能会找到一个比您自己提出的更简单、不那么独特、更易于维护的ML解决方案。多年来，我参与的最成功的项目都来自于与中小企业(sme)组成的广泛团队(如果我幸运的话，还有实际的最终用户)进行这种创造性的讨论，这让我能够将自己的想法转变为创造性的方式，尽可能地接近他们的愿景。

但是，讨论理想的最终状态并不仅仅是为了获得更令人惊叹的ML解决方案。让项目发起人参与进来，可以让他们的观点、想法和创造力以积极的方式影响项目。讨论还有助于在项目开发中建立信任和归属感，这有助于将团队团结在一起。

学习密切倾听机器学习项目客户的需求是机器学习工程师最重要的技能之一，远远超过掌握任何算法、语言或平台。它将帮助指导你要尝试什么，你要研究什么，以及如何以不同的方式思考问题，从而尽可能地提出最佳解决方案。

在图4.4所示的场景中，最初的规划会议产生了理想状态的粗略草图。这可能不会是最终的引擎(根据我的经验，这是绝对不可能的)。但是这个图将告诉我们如何将这些功能块转换成系统。它将有助于告知实验的方向，以及您和团队需要彻底研究的项目领域，以最小化或防止意外的范围蔓延，如图4.5所示。

计算机代写|机器学习代写Machine Learning代考|Scope creep: An almost guaranteed assassination of a project

不恰当的计划(或在没有涉及项目团队的情况下进行计划)是导致项目缓慢死亡的最令人沮丧的方法之一。这个概念也被称为“千次请求导致机器学习死亡”，它在开发的后期阶段实现，特别是当向不了解项目构建细节的团队展示演示时。如果客户(内部业务单位)不是计划讨论的一方，他们不可避免地会有很多问题，关于演示的功能。
在我所见过的几乎每一个案例中(或者在我早期尝试在不要求输入的情况下“英雄般地完成”一个项目)，演示会议的结果将是许多附加功能和需求的添加请求。这是预期的(即使在一个适当设计和计划的项目中)，但是如果实现不能轻松地包括与不可变的业务操作“定律”相关的关键特性，则可能需要对项目进行完全的重新实现。这让决策者面临一个艰难的选择:是因为DS团队(或个人)做出的决定而推迟项目，还是为了防止最初的失败重演而完全放弃项目。
在ML世界中，没有什么比在产品投入使用后立即听到强烈的负面反馈更具有破坏性的了。收到大量高管发来的电子邮件，表示你刚刚推出的解决方案是向狗主人推荐猫玩具，这很可笑，但向儿童推荐成人主题产品，这几乎是最糟糕的了。唯一更糟糕的是，在项目交付之前，在用户验收测试(UAT)期间，发现需要做出一个无法克服的更改列表来满足业务的紧急需求，并且从头开始项目比对现有解决方案进行更改花费的时间更少。
识别范围蔓延是很重要的，但它的大小可以最小化，在某些情况下可以消除。在实验笔记本或IDE中输入单个字符之前，需要达到适当的讨论级别，并将项目的关键方面包含在有时令人痛苦的递归和痛苦的细节中。

计算机代写|机器学习代写Machine Learning代考请认准UprivateTA™. UprivateTA™为您的留学生涯保驾护航。

微观经济学代写

线性代数代写

线性代数是数学的一个分支，涉及线性方程，如：线性图，如：以及它们在向量空间和通过矩阵的表示。线性代数是几乎所有数学领域的核心。

博弈论代写

微积分代写

计量经济学代写

MATLAB代写

2023年5月18日2023年5月18日 CS代写, 数据库, 计算机代写

计算机代写|数据库代考Database代考|TABLE, ROWS, AND COLUMNS

如果你也在怎样代写数据库Database 这个学科遇到相关的难题，请随时右上角联系我们的24/7代写客服。数据库Database在计算机领域，数据库是一个有组织的数据集合，以电子方式存储和访问。小型数据库可以存储在文件系统中，而大型数据库则托管在计算机集群或云存储中。数据库的设计跨越了形式技术和实际考虑，包括数据建模、有效的数据表示和存储、查询语言、敏感数据的安全和隐私，以及分布式计算问题，包括支持并发访问和容错。

数据库Database数据库管理系统（DBMS）是与终端用户、应用程序和数据库本身交互的软件，用于捕获和分析数据。DBMS软件还包括了为管理数据库而提供的核心设施。数据库、DBMS和相关应用程序的总和可以被称为数据库系统。通常，术语 “数据库 “也被宽泛地用来指代任何一个DBMS、数据库系统或与数据库相关的应用程序。

数据库Database代写，免费提交作业要求，满意后付款，成绩80\%以下全额退款，安全省心无顾虑。专业硕博写手团队，所有订单可靠准时，保证 100% 原创。最高质量的数据库Database作业代写，服务覆盖北美、欧洲、澳洲等国家。在代写价格方面，考虑到同学们的经济条件，在保障代写质量的前提下，我们为客户提供最合理的价格。由于作业种类很多，同时其中的大部分作业在字数上都没有具体要求，因此数据库Database作业代写的价格不固定。通常在专家查看完作业要求之后会给出报价。作业难度和截止日期对价格也有很大的影响。

avatest™帮您通过考试

•最快12小时交付

•200+ 英语母语导师

•70分以下全额退款

想知道您作业确定的价格吗? 免费下单以相关学科的专家能了解具体的要求之后在1-3个小时就提出价格。专家的报价比上列的价格能便宜好几倍。

我们在计算机Quantum computer代写方面已经树立了自己的口碑, 保证靠谱, 高质且原创的计算机Quantum computer代写服务。我们的专家在数据库Database代写方面经验极为丰富，各种数据库Database相关的作业也就用不着说。

计算机代写|数据库代考Database代考|TABLE, ROWS, AND COLUMNS

Informally you can think of a relational database as a collection of tables, with each containing rows and columns. At this level, it looks a lot like a computerized workbook containing several worksheets (or spreadsheets), although a worksheet is much less constrained than a database table is. You can put just about anything in any cell in a worksheet. In contrast, every entry in a particular column of a table is expected to contain the same kind of data. For example, all of the cells in a particular column might contain phone numbers or last names.
NOTE Actually, a poorly designed database application may allow the user to sneak some strange kinds of data into fields, too. For example, if the database and user interface aren’t designed properly, you might be able to enter a string such as “none” in a telephone number field. That’s not the field’s intent, however. In contrast, a spreadsheet’s cells don’t really care what you put in them.
The set of the values that are allowed for a column is called the column’s domain. For example, a column’s domain might be telephone numbers, bank account numbers, snowshoe sizes, or hang glider colors.
Domain is closely related to data type, but it’s not quite the same. A column’s data type is the kind of data that the column can hold. (Yes, I know this is pretty obvious.) The data types that you can use for a column depend on the particular database you are using, but typical data types include integer, floating-point number (a number with a decimal point), string, and date.

To see the difference between domain and data type, note that street address (323 Relational Rd) and jersey color (red) are both strings. However, the domain for the street address column is valid street addresses, whereas the domain for the jersey color column is colors (and possibly not even all colors if you only allow a few choices). You can think of the data type as the highest level or most general possible domain. For example, an address or color domain is a more restrictive subset of the domain allowing all strings.
The rows in a table correspond to column values that are related to each other according to the table’s purpose. For example, suppose you have a Competitors table that contains contact information for participants in your First (and probably Last) Annual Extreme Pyramid Sports Championship (aka the Cairolympics). This table includes columns to hold competitor name, address, event, blood type, and next of kin, as shown in Figure 2.1. (Note that this is not a good database design. You’ll see why in later chapters.)

计算机代写|数据库代考Database代考|RELATIONS, ATTRIBUTES, AND TUPLES

The values in a row are related by the fact that they apply to a particular person. Because of this fact, the formal term for a table is a relation. This may cause some confusion because the word relation is also used informally to describe a relationship between two tables. This use is described in the section “Foreign Key Constraints,” later in this chapter.
The formal term for a column is an attribute or data element. For example, in the Competitors relation shown in Figure 2.1, Name, Address, BloodType, and NextOfKin are the attributes of each of the people represented. You can think of this as in “each person in the relation has a Name attribute.”
The formal term for a row is a tuple (rhymes with “scruple”). This almost makes sense if you think of a two-attribute relation as holding data pairs, a three-attribute relation as holding value triples, and a four-attribute relation as holding data quadruples. Beyond four items, mathematicians would say 5 -tuple, 6-tuple, and so forth, hence the name tuple.
Don’t confuse the formal term relation (meaning table) with the more general and less precise use of the term that means “related to” as in “these fields form a relation between these two tables” (or “that psycho is no relation of mine”). Similarly, don’t confuse the formal term attribute with the less precise use that means “feature of” as in “this field has the ‘required’ attribute” (or “don’t attribute that comment to me!”). I doubt you’ll confuse the term tuple with anything-it’s probably confusing enough all by itself.

Theoretically a relation does not impose any ordering on the tuples that it contains, nor does it give an ordering to its attributes. Generally, the orderings don’t matter to mathematical database theory. In practice, however, database applications usually sort the records selected from a table in some manner to make it easier for the user to understand the results. It’s also a lot easier to write the program (and for the user to understand) if the order of the fields remains constant, so database products typically return fields in the order in which they were created in the table unless told otherwise.

数据库代写

计算机代写|数据库代考Database代考|TABLE, ROWS, AND COLUMNS

非正式地，您可以将关系数据库视为表的集合，每个表包含行和列。在这个级别上，它看起来很像一个包含几个工作表(或电子表格)的计算机化工作簿，尽管工作表比数据库表受的约束要少得多。您可以在工作表的任何单元格中放入任何内容。相反，表中特定列中的每个条目都应该包含相同类型的数据。例如，特定列中的所有单元格可能包含电话号码或姓氏。
实际上，设计糟糕的数据库应用程序也可能允许用户在字段中插入一些奇怪的数据。例如，如果数据库和用户界面设计不当，您可能会在电话号码字段中输入“none”之类的字符串。然而，这并不是该领域的意图。相反，电子表格的单元格并不关心你在里面放了什么。
列允许的值的集合称为列的域。例如，列的域可能是电话号码、银行帐号、雪鞋尺寸或滑翔机颜色。
域与数据类型密切相关，但并不完全相同。列的数据类型是该列可以保存的数据类型。(是的，我知道这是显而易见的。)可用于列的数据类型取决于所使用的特定数据库，但典型的数据类型包括整数、浮点数(带有小数点的数字)、字符串和日期。

要查看域和数据类型之间的区别，请注意街道地址(323 Relational Rd)和球衣颜色(红色)都是字符串。但是，街道地址列的域是有效的街道地址，而球衣颜色列的域是颜色(如果只允许几种选择，甚至可能不是所有颜色)。您可以将数据类型视为最高级或最通用的域。例如，地址域或颜色域是允许所有字符串的域的更严格的子集。
表中的行对应于根据表的用途相互关联的列值。例如，假设您有一个参赛者表，其中包含第一届(也可能是最后一届)年度极限金字塔体育锦标赛(又名开罗奥运会)参与者的联系信息。该表包括用于保存参赛者姓名、地址、事件、血型和近亲的列，如图2.1所示。(注意，这不是一个好的数据库设计。你会在后面的章节中看到原因。)

计算机代写|数据库代考Database代考|RELATIONS, ATTRIBUTES, AND TUPLES

一行中的值是通过它们适用于特定的人这一事实相关联的。由于这个事实，表的正式术语是关系。这可能会引起一些混淆，因为“关系”这个词也非正式地用于描述两个表之间的关系。这种用法将在本章后面的“外键约束”一节中描述。
列的正式术语是属性或数据元素。例如，在图2.1所示的Competitors关系中，Name、Address、blood – type和NextOfKin是所表示的每个人的属性。您可以将其理解为“关系中的每个人都有一个Name属性”。
行的正式术语是元组(与“scruple”押韵)。如果将双属性关系视为保存数据对，将三属性关系视为保存值三元组，将四属性关系视为保存数据三元组，那么这几乎是有意义的。超过4个元素，数学家会说5 -tuple, 6-tuple，以此类推，因此称为元组。
不要混淆正式的术语关系(意思是表)和更一般、更不精确的术语“与…相关”的用法，比如“这些字段形成了这两个表之间的关系”(或者“那个心理不是我的关系”)。类似地，不要将正式术语attribute与表示“特性”的不太精确的用法混淆，如“此字段具有’required’属性”(或“不要将该注释归因于我!”)。我怀疑您会将术语元组与任何东西混淆——它本身可能就足够令人困惑了。

理论上，关系不会对它所包含的元组施加任何排序，也不会对其属性给出排序。一般来说，排序与数学数据库理论无关。但是，在实践中，数据库应用程序通常以某种方式对从表中选择的记录进行排序，以便用户更容易理解结果。如果字段的顺序保持不变，编写程序也会容易得多(用户也容易理解)，因此数据库产品通常按照在表中创建字段的顺序返回字段，除非另有说明。

计算机代写|数据库代考Database代考请认准UprivateTA™. UprivateTA™为您的留学生涯保驾护航。

微观经济学代写

线性代数代写

线性代数是数学的一个分支，涉及线性方程，如：线性图，如：以及它们在向量空间和通过矩阵的表示。线性代数是几乎所有数学领域的核心。

博弈论代写

微积分代写

计量经济学代写

MATLAB代写

2023年5月18日2023年5月18日 CS代写, 数据库, 计算机代写

计算机代写|数据库代考Database代考|PICKING A DATABASE

avatest™帮您通过考试

•最快12小时交付

•200+ 英语母语导师

•70分以下全额退款

想知道您作业确定的价格吗? 免费下单以相关学科的专家能了解具体的要求之后在1-3个小时就提出价格。专家的报价比上列的价格能便宜好几倍。

计算机代写|数据库代考Database代考|PICKING A DATABASE

There’s an expression, “If all you have is a hammer, everything looks like a nail.” If the only kind of database you understand is the relational database, then you’ll probably try to hammer every kind of data into a relational database, and that can sometimes lead to trouble.

I once worked on a fairly large database application with 40 developers and more than 120,000 lines of code. The program loaded some fairly large relational databases and used their data to build huge, tree-like structures. Those structures allowed sales representatives to design and modify extremely complicated projects for customers involving tens of thousands of line items.
The data was naturally hierarchical but was stored in relational databases, so the program was forced to spend a long time loading each data set. Many projects took 5 to 20 minutes to load. When the user made even a simple change, the program’s design required it to recalculate parts of the tree and then save the changes back into the database-a process that took another 5 to 30 minutes depending on the complexity of the model. The program was so slow that the users couldn’t perform the types of experiments they needed to optimize the projects they were building. You couldn’t quickly see the effects of tweaking a couple of numbers here and there.
To make matters worse, loading and saving all of that hierarchical data in a relational database required tens of thousands of lines of moderately tricky code that was hard to debug and maintain.
At one point, I performed a quick experiment to see what would happen if the data were stored in an XML database, a database that naturally stores hierarchical data. My test program was able to load and save data sets containing 20,000 items in 3 to 4 seconds.
At that point, the project was too big and the design too entrenched to make such a fundamental change. (Afterward, political pressure within the company pulled the project in too many directions and it eventually shredded like a tissue in a tug-of-war.)

The lesson is clear: before you spend a lot of time building the ultimate relational database and piling thousands of lines of code on top of it, make sure that’s the kind of database you need. Had this project started with an XML database, it probably would have had a simpler, more natural design with much less code and would probably have lasted for many years to come.

计算机代写|数据库代考Database代考|RELATIONAL POINTS OF VIEW

Relational databases play a critical role in many important (i.e., money-related) computer applications. As is the case whenever enormous amounts of money are at stake, people have spent a huge amount of time and effort building, studying, and refining relational databases. Database researchers usually approach relational databases from one of three points of view.

The first group approaches the problem from a database-theoretical point of view. These people tend to think in terms of provability, mathematical set theory, and propositional logic. You’ll see them at parties throwing around phrases like relational algebra, Cartesian product, and tuple relational calculus. This approach is intellectually stimulating (and looks good on a résumé) but can be a bit intimidating. These researchers focus on logical design and idealized database principles.
The second group approaches the matter from a less formal “just build the database and get it done” point of view. Their terminology tends to be less precise and rigorous but more intuitive. They tend to use terms that you may have heard before like table, row, and column. These people focus on physical database design and pay more attention to concrete bits-and-bytes issues dealing with actually building a database and getting the most out of it.

The third group tends to think in terms of flat files and the underlying disk structure used to hold data. Although these people are probably in the minority these days, their terms file, record, and field snuck into database nomenclature and stuck. Many who still use these terms are programmers and other developers who look at the database from a consumer’s “how do I get my data out of it” point of view.

These differing viewpoints have led to several different and potentially puzzling ways to consider relational databases. This can cause some confusion, particularly because the different groups have latched on to some of the same terms but used them for different meanings. In fact, they sometimes use the term relation in very different ways (that are described later in this chapter).

This chapter loosely groups these terms into “formal” and “informal” categories, where the formal category includes the database theoretical terms and the informal category includes everything else.
This chapter begins with informal terms. Each section initially focuses on informal terms and concepts, and then explains how they fit together with their more formal equivalents.

数据库代写

计算机代写|数据库代考Database代考|PICKING A DATABASE

有句话是这样说的:“如果你只有一把锤子，那么所有的东西看起来都像钉子。”如果您唯一了解的数据库类型是关系数据库，那么您可能会尝试将所有类型的数据都放入关系数据库中，这有时会导致麻烦。

我曾经开发过一个相当大的数据库应用程序，有40名开发人员和超过12万行代码。该程序加载了一些相当大的关系数据库，并使用它们的数据构建巨大的树状结构。这些结构允许销售代表为涉及成千上万行项目的客户设计和修改极其复杂的项目。
数据自然是分层的，但存储在关系数据库中，因此程序不得不花费很长时间加载每个数据集。许多项目需要5到20分钟来加载。当用户做了一个简单的更改时，程序的设计要求它重新计算树的部分，然后将更改保存回数据库-这个过程又需要5到30分钟，具体取决于模型的复杂程度。这个程序非常缓慢，以至于用户无法执行他们需要的实验类型来优化他们正在构建的项目。你不能很快看到这里和那里调整几个数字的效果。
更糟糕的是，在关系数据库中加载和保存所有分层数据需要数万行比较复杂的代码，这些代码很难调试和维护。
有一次，我执行了一个快速实验，看看如果将数据存储在XML数据库中会发生什么情况，XML数据库是一种自然存储分层数据的数据库。我的测试程序能够在3到4秒内加载并保存包含20,000项的数据集。
在那个时候，这个项目太大了，设计也太根深蒂固了，无法做出这样的根本改变。(后来，公司内部的政治压力把这个项目扯向了太多的方向，最终它像拔河中的纸巾一样被撕成了碎片。)

教训很清楚:在花费大量时间构建最终的关系数据库并在其上堆积数千行代码之前，请确保这是您需要的数据库类型。如果这个项目从XML数据库开始，它可能会有一个更简单、更自然的设计和更少的代码，并且可能会持续许多年。

计算机代写|数据库代考Database代考|RELATIONAL POINTS OF VIEW

关系数据库在许多重要的(例如，与金钱有关的)计算机应用程序中起着关键作用。当大量的金钱面临风险时，人们花费了大量的时间和精力来构建、研究和改进关系数据库。数据库研究人员通常从以下三种观点之一来研究关系数据库。

第一组从数据库理论的角度来解决这个问题。这些人倾向于用可证明性、数学集合论和命题逻辑来思考。你会在聚会上看到他们抛出关系代数、笛卡尔积和元组关系演算之类的短语。这种方法在智力上是刺激的(而且看起来很不错)，但可能有点吓人。这些研究人员专注于逻辑设计和理想化的数据库原则。
第二组从一种不那么正式的“只要构建数据库并完成它”的观点来处理这个问题。他们的术语往往不那么精确和严谨，但更直观。他们倾向于使用您可能之前听过的术语，如表、行和列。这些人专注于物理数据库设计，更关注具体的位和字节问题，以处理实际构建数据库并充分利用它。

第三组倾向于考虑平面文件和用于保存数据的底层磁盘结构。尽管这些人现在可能是少数，但他们的术语文件、记录和字段悄悄进入了数据库的命名法，并被卡住了。许多仍在使用这些术语的人是程序员和其他开发人员，他们从消费者“如何从数据库中获取数据”的角度来看待数据库。

这些不同的观点导致了考虑关系数据库的几种不同的、可能令人困惑的方式。这可能会引起一些混淆，特别是因为不同的群体已经锁定了一些相同的术语，但使用它们的含义不同。事实上，它们有时以非常不同的方式使用术语关系(这将在本章后面描述)。

本章将这些术语粗略地分为“正式”和“非正式”两类，其中正式类别包括数据库理论术语，非正式类别包括其他所有术语。
本章从非正式用语开始。每个部分首先关注非正式术语和概念，然后解释它们如何与更正式的对等物结合在一起。

计算机代写|数据库代考Database代考请认准UprivateTA™. UprivateTA™为您的留学生涯保驾护航。

微观经济学代写

线性代数代写

线性代数是数学的一个分支，涉及线性方程，如：线性图，如：以及它们在向量空间和通过矩阵的表示。线性代数是几乎所有数学领域的核心。

博弈论代写

微积分代写

计量经济学代写

MATLAB代写

2023年5月18日2023年5月18日 CS代写, 数据库, 计算机代写

计算机代写|数据库代考Database代考|Retrieval

avatest™帮您通过考试

•最快12小时交付

•200+ 英语母语导师

•70分以下全额退款

想知道您作业确定的价格吗? 免费下单以相关学科的专家能了解具体的要求之后在1-3个小时就提出价格。专家的报价比上列的价格能便宜好几倍。

计算机代写|数据库代考Database代考|Retrieval

Retrieval, which is another word for read, is the R in CRUD. (Luckily, “retrieve” also starts with an R, so you don’t need to memorize a new acronym.) Your database should allow you to find every piece of data. There is no point in putting something in the database if it cannot be retrieved later. (That would be a “data black hole,” not a database.)

A database should allow you to build data so that a particular chunk of data can be found in one or more specific ways. For example, you should be able to find a customer’s billing record by searching for the customer name or customer ID.

Ideally, the database will also allow you to build your data so that it is relatively quick and easy to retrieve in specific ways.
For example, suppose you want to see where your customers live to determine whether you should start a delivery service in a new city. In order to get this information, it would be helpful to be able to locate customers based on their address. Ideally, you can optimize your database structure so that you can quickly search for customers by address.

Conversely, you may not need to search for customers by middle name as often. (Imagine a customer calls you and says, “Can you check my records?” I don’t remember whether I paid the bill last month. I don’t remember my account number or last name either, but my middle name is Konfused.” It would be nice if you could search by address faster than by middle name.

Being able to find all the data in a database quickly and reliably is an important part of database design. Finding the data you need in a poorly designed database can take hours or days, not seconds.

计算机代写|数据库代考Database代考|Consistency

Another aspect of $\ mathm {R}$in CRUD is consistency. (It is pure coincidence that both “consistency” and CRUD begin with $\ mathm {C}$. Don’t get too excited.) The database should provide consistent results. If you perform the same search twice in a row, you should get the same results. Another user performing the same search should get the same results. (Of course, this assumes that the underlying data hasn’t changed in the meantime. You can’t expect your net worth query to get the same results every day when stock prices are moving wildly.)

A well-built database product can ensure that the exact same queries return the same results, but design also plays an important role. If the database is poorly designed, conflicting data may be stored in different parts of the database. For example, you can store one set of contact information in a customer’s order and another set of information in a master customer record. Later, if you need to contact a customer with a question about an order, what contact information should you use?

Validity is closely related to the concept of consistency. Consistency means that different parts of the database don’t have conflicting views about the same information. Validity means checking other data in the database where possible. In CRUD terms, data can be validated when a record is created, updated, or deleted.

Just like physical data containers, computerized databases can hold incomplete, incorrect, or contradictory data. You can never protect a database from users who can’t spell or simply type error information, but good database design can help prevent certain types of errors that a physical database cannot.

For example, the database can easily verify that the data has the correct type. If the user sees the Date field and enters “No thanks, I’m married,” the database knows that this is not a valid date format and rejects the value. Similarly, it can determine that “Old” is not a valid Age, “Lots” is not a valid Quantity, and 3 is not a valid PhoneNumber.

The database can also verify that the value entered by the user exists in another part of the database. For example, a poor typist trying to type $\ mathm {CO}$into the status field might type $\ mathm {CP}$. The database can check the list of valid states and reject the data if no CP list is found. The user interface can also avoid this problem by letting the user select state from a drop-down list, but the database should still protect itself from invalid data as much as possible.

数据库代写

计算机代写|数据库代考Database代考|Retrieval

检索是“读取”的另一个词，是CRUD中的R。(幸运的是，“检索”也是以R开头的，所以你不需要记住一个新的缩写词。)您的数据库应该允许您找到每一条数据。如果以后无法检索，那么将某些内容放入数据库就没有意义了。(那将是一个“数据黑洞”，而不是数据库。)

数据库应该允许您构建数据，以便能够以一种或多种特定的方式找到特定的数据块。例如，您应该能够通过搜索客户名称或客户ID来查找客户的账单记录。

理想情况下，数据库还将允许您构建数据，以便以特定方式相对快速和容易地获取数据。
例如，假设您想查看客户的居住地，以便决定是否应该在一个新城市开展送货服务。为了获得这些信息，能够根据客户的地址找到他们将是很有帮助的。理想情况下，您可以优化数据库结构，以便可以按地址快速搜索客户。

相反，您可能不需要太频繁地通过中间名搜索客户。(想象一下，一个顾客打电话给你说:“你能查一下我的记录吗?”我不记得上个月是否付过帐。我也不记得我的账号和姓氏，但我的中间名是“糊涂”(Konfused)。”)如果按地址搜索的速度比按中间名搜索的速度快就好了。

能够快速、可靠地找到数据库中的所有数据是数据库设计的一个重要部分。在设计糟糕的数据库中找到所需的数据可能需要数小时或数天，而不是几秒钟。

计算机代写|数据库代考Database代考|Consistency

CRUD中$\ mathm {R}$的另一个方面是一致性。(“一致性”和CRUD都以$\ mathm {C}$开头，这纯粹是巧合。不要太兴奋。)数据库应该提供一致的结果。如果在一行中执行两次相同的搜索，应该会得到相同的结果。执行相同搜索的另一个用户也应该得到相同的结果。(当然，这假定底层数据在此期间没有改变。当股价剧烈波动时，你不能指望你的净值查询每天都得到同样的结果。)

构建良好的数据库产品可以确保完全相同的查询返回相同的结果，但设计也起着重要作用。如果数据库设计得很差，可能会在数据库的不同部分存储冲突的数据。例如，您可以在客户的订单中存储一组联系人信息，在主客户记录中存储另一组信息。稍后，如果您需要与客户联系有关订单的问题，您应该使用哪些联系信息?

效度与一致性的概念密切相关。一致性意味着数据库的不同部分不会对相同的信息持有矛盾的观点。有效性意味着在可能的情况下对数据库中的其他数据进行检查。在CRUD术语中，可以在创建、更新或删除记录时验证数据。
就像物理数据容器一样，计算机化的数据库可以保存不完整、不正确或矛盾的数据。您永远无法保护数据库免受不会拼写或只是简单地输入错误信息的用户的影响，但是良好的数据库设计可以帮助防止物理数据库无法防止的某些类型的错误。

例如，数据库可以很容易地验证数据是否具有正确的类型。如果用户看到Date字段并输入“No thanks, I’m married”，数据库就会知道这不是有效的日期格式，并拒绝接受该值。类似地，它可以判断“Old”不是有效的Age，“Lots”不是有效的Quantity, 3不是有效的PhoneNumber。
数据库还可以验证用户输入的值是否存在于数据库的另一部分中。例如，一个蹩脚的打字员试图在状态字段中输入$\ mathm {CO}$，可能会键入$\ mathm {CP}$。数据库可以检查有效状态列表，如果没有找到CP列表，则拒绝接受数据。用户界面也可以让用户从下拉列表中选择状态，从而避免这个问题，但是数据库仍然应该尽可能保护自己免受无效数据的侵害。

计算机代写|数据库代考Database代考请认准UprivateTA™. UprivateTA™为您的留学生涯保驾护航。

微观经济学代写

线性代数代写

线性代数是数学的一个分支，涉及线性方程，如：线性图，如：以及它们在向量空间和通过矩阵的表示。线性代数是几乎所有数学领域的核心。

博弈论代写

微积分代写

计量经济学代写

MATLAB代写

2023年5月16日2023年5月16日 CS代写, Neural Networks, 神经网络, 计算机代写

计算机代写|神经网络代写Neural Networks代考|RPROP Arguments

如果你也在怎样代写神经网络Neural Networks AIML425这个学科遇到相关的难题，请随时右上角联系我们的24/7代写客服。神经网络Neural Networks理论既有助于更好地确定大脑中的神经元如何运作，也为创造人工智能的努力提供了基础。反映了人脑的行为，使计算机程序能够识别模式并解决人工智能、机器学习和深度学习领域的常见问题。

神经网络Neural Networks也被称为人工神经网络（ANN）或模拟神经网络（SNN），是机器学习的一个子集，是深度学习算法的核心。它们的名称和结构受到人脑的启发，模仿了生物神经元相互之间的信号方式。

神经网络Neural Networks作业代写，免费提交作业要求，满意后付款，成绩80\%以下全额退款，安全省心无顾虑。专业硕博写手团队，所有订单可靠准时，保证 100% 原创。my-assignmentexpert™，最高质量的神经网络Neural Networks作业代写，服务覆盖北美、欧洲、澳洲等国家。在代写价格方面，考虑到同学们的经济条件，在保障代写质量的前提下，我们为客户提供最合理的价格。由于统计Statistics作业种类很多，同时其中的大部分作业在字数上都没有具体要求，因此神经网络Neural Networks作业代写的价格不固定。通常在经济学专家查看完作业要求之后会给出报价。作业难度和截止日期对价格也有很大的影响。

avatest™帮您通过考试

•最快12小时交付

•200+ 英语母语导师

•70分以下全额退款

想知道您作业确定的价格吗? 免费下单以相关学科的专家能了解具体的要求之后在1-3个小时就提出价格。专家的报价比上列的价格能便宜好几倍。

计算机代写|神经网络代写Neural Networks代考|RPROP Arguments

As previously mentioned, one advantage RPROP has over backpropagation is that no training arguments need to be provided in order for RPROP to be used. That is not to say that there are no configuration settings for RPROP. The configuration settings for RPROP do not usually need to be changed from their defaults. However, if you really want to change them, there are several configuration settings that you can set for RPROP training. These configuration settings are:

Initial Update Values
Maximum Step
As you will see in the next section, RPROP keeps an array of update values for the weights. This determines how large of a change will be made to each weight. This is something like the learning rate in backpropagation, only much better. There is an update value for every weight in the neural network. This allows the update values to be fine tuned to each individual weight as training progresses. Some backpropagation algorithms will vary the learning rate and momentum as learning progresses. The RPROP approach is much better, because unlike backpropagation, it does not simply use a single learning rate for the entire neural network.

These update values must start from somewhere. The “initial update values” argument defines this. By default, this argument is set to a value of 0.1 . As a general rule, this default should never be changed. One possible exception to this is in a neural network that has already been trained. If the neural network is already trained, then some of the initial update values are going to be too strong for the neural network. The neural network will regress for many iterations before it is able to improve. An already trained neural network may benefit from a much smaller initial update.

计算机代写|神经网络代写Neural Networks代考|Understanding RPROP

In the last few sections, the arguments, constants and data structures necessary for RPROP were covered. In this section, we will see exactly how to run through an iteration of RPROP. In the next section, we will apply real numbers to RPROP and see how training iterations progress for an XOR training. We will train exactly the same network that we used with backpropagation. This will give us a good idea of the difference in performance of backpropagation compared to RPROP.

When we talked about backpropagation, we mentioned two weight update methods: online and batch. RPROP does not support online training. All weight updates used with RPROP will be performed in batch mode. Because of this, each iteration of RPROP will receive gradients that are the sum of the individual gradients of each training set. This is consistent with using backpropagation in batch mode.

There are three distinct steps in an iteration of an RPROP iteration. They are covered in the next three sections.

Determine Sign Change of Gradient
At this point, we should have the gradients. These gradients are nearly exactly the same as the gradients calculated by the backpropagation algorithm. The only difference is that RPROP uses a gradient that is the inverse of the backpropagation gradient. This is easy enough to adjust. Simply place a negative operator in front of every backpropagation gradient. Because the same process is used to obtain gradients in both RPROP and backpropagation, we will not repeat it here. To learn how to calculate a gradient, refer to Chapter 4.
The first step is to compare the gradient of the current iteration to the gradient of the previous iteration. If there is no previous iteration, then we can assume that the previous gradient was zero.

To determine whether the gradient sign has changed, we will use the sign (sgn) function. The sgn function is defined in Equation 5.1.
Equation 5.1: The Sign Function (sgn)
$$
\operatorname{sgn}(x)= \begin{cases}-1 & \text { if } x<0 \\ 0 & \text { if } x=0 \\ 1 & \text { if } x>0\end{cases}
$$

神经网络代写

计算机代写|神经网络代写Neural Networks代考|RPROP Arguments

如前所述，RPROP相对于反向传播的一个优点是，使用RPROP不需要提供训练参数。这并不是说没有RPROP的配置设置。RPROP的配置设置通常不需要更改其默认值。但是，如果您真的想更改它们，您可以为RPROP培训设置几个配置设置。这些配置设置是:

初始更新值

最大步长
在下一节中您将看到，RPROP为权重保存了一个更新值数组。这决定了将对每个权重进行多大的更改。这有点像反向传播中的学习率，只是要好得多。神经网络中的每个权重都有一个更新值。这使得更新值可以随着训练的进行微调到每个单独的权重。一些反向传播算法会随着学习的进展而改变学习速率和动量。RPROP方法要好得多，因为与反向传播不同，它不只是对整个神经网络使用单一的学习率。

这些更新值必须从某个地方开始。“initial update values”参数定义了这一点。默认情况下，该参数被设置为0.1。作为一般规则，这个默认值永远不应该更改。一个可能的例外是已经训练过的神经网络。如果神经网络已经被训练过了，那么一些初始更新值对于神经网络来说会太强。神经网络在能够改进之前会进行多次迭代回归。一个已经训练好的神经网络可能会从一个小得多的初始更新中受益。

计算机代写|神经网络代写Neural Networks代考|Understanding RPROP

在最后几节中，介绍了RPROP所需的参数、常量和数据结构。在本节中，我们将确切地看到如何运行RPROP的迭代。在下一节中，我们将对RPROP应用实数，并查看XOR训练的训练迭代是如何进行的。我们将训练与反向传播完全相同的网络。这将使我们很好地了解反向传播与RPROP在性能上的差异。

当我们讨论反向传播时，我们提到了两种权重更新方法:在线和批处理。RPROP不支持在线培训。与RPROP一起使用的所有权重更新都将以批处理模式执行。因此，RPROP的每次迭代将接收到的梯度是每个训练集的单个梯度的总和。这与在批处理模式下使用反向传播是一致的。

在RPROP迭代的迭代中有三个不同的步骤。它们将在接下来的三节中介绍。

确定梯度的符号变化
现在，我们应该有梯度了。这些梯度与反向传播算法计算的梯度几乎完全相同。唯一的区别是RPROP使用的梯度与反向传播梯度相反。这很容易调整。只需在每个反向传播梯度前面放置一个负运算符。因为在RPROP和反向传播中使用相同的过程来获得梯度，所以我们在这里不再重复。要学习如何计算梯度，请参阅第4章。
第一步是将当前迭代的梯度与之前迭代的梯度进行比较。如果没有之前的迭代，那么我们可以假设之前的梯度为零。

为了确定梯度符号是否发生了变化，我们将使用sign (sgn)函数。sgn函数定义如式5.1所示。
方程5.1:符号函数(sgn)
$$
\operatorname{sgn}(x)= \begin{cases}-1 & \text { if } x<0 \\ 0 & \text { if } x=0 \\ 1 & \text { if } x>0\end{cases}
$$

计算机代写|机器学习代写Machine Learning代考请认准UprivateTA™. UprivateTA™为您的留学生涯保驾护航。

微观经济学代写

线性代数代写

线性代数是数学的一个分支，涉及线性方程，如：线性图，如：以及它们在向量空间和通过矩阵的表示。线性代数是几乎所有数学领域的核心。

博弈论代写

微积分代写

计量经济学代写

MATLAB代写

2023年5月16日2023年5月16日 CS代写, Neural Networks, 神经网络, 计算机代写

计算机代写|神经网络代写Neural Networks代考|Using the Chain Rule

avatest™帮您通过考试

•最快12小时交付

•200+ 英语母语导师

•70分以下全额退款

想知道您作业确定的价格吗? 免费下单以相关学科的专家能了解具体的要求之后在1-3个小时就提出价格。专家的报价比上列的价格能便宜好几倍。

计算机代写|神经网络代写Neural Networks代考|Using the Chain Rule

There are many different rules in Calculus to allow you to take derivatives manually. We just saw an example of the power rule. This rule states that given the equation:
$$
f(x)=x^n
$$
the derivative of $\mathrm{f}(\mathrm{x})$ will be as follows:
$$
f^{\prime}(x)=n x^{n-1}
$$
This allows you to quickly take the derivative of any power. There are many other derivative rules, and they are very useful to know. However, if you do not wish to learn manual differentiation, you can generally get by without it by using a program such as $R$.

However, there is one more rule that is very useful to know. This rule is called the chain rule. The chain rule deals with composite functions. A composite function is nothing more than when one function takes the results of a second function as input. This may sound complex, but programmers make use of composite functions all the time. Here is an example of a composite function call in Java.
System.out.printin( Math.pow $(3,2)$ );
This is a composite function because we take the result of the function pow and feed it to println.

计算机代写|神经网络代写Neural Networks代考|Understanding Gradients

The first step is to calculate the gradients of the neural network. The gradients are used to calculate the slope, or gradient, of the error function for a particular weight. A weight is a connection between two neurons. Calculating the gradient of the error function allows the training method to know that it should either increase or decrease the weight. There are a number of different training methods that make use of gradients. These training methods are called propagation training. This book will discuss the following propagation training methods:

Backpropagation
Resilient Propagation
Quick Propagation
This chapter will focus on using the gradients to train the neural network using backpropagation. The next few chapters will cover the other propagation methods.
What is a Gradient
First of all, let’s look at what a gradient is. Basically, training is a search. You are searching for the set of weights that will cause the neural network to have the lowest global error for a training set. If we had an infinite amount of computation resources, we would simply try every possible combination of weights and see which one provided the absolute best global error.
Because we do not have unlimited computing resources, we have to use some sort of shortcut. Essentially, all neural network training methods are really a kind of shortcut. Each training method is a clever way of finding an optimal set of weights without doing an impossibly exhaustive search.
Consider a chart that shows the global error of a neural network for each possible weight. This graph might look something like Figure 4.1 .

神经网络代写

计算机代写|神经网络代写Neural Networks代考|Using the Chain Rule

微积分中有很多不同的规则可以让你手动求导数。我们刚刚看到了幂法则的一个例子。该规则指出，给定方程:
$$
f(x)=x^n
$$
$\mathrm{f}(\mathrm{x})$的导数为:
$$
f^{\prime}(x)=n x^{n-1}
$$
这样就可以快速求任意次幂的导数。还有很多其他的导数法则，它们都很有用。但是，如果您不希望学习手动区分，通常可以通过使用$R$之类的程序来完成。

然而，还有一条规则是非常有用的。这个法则叫做链式法则。链式法则处理复合函数。复合函数只不过是一个函数将另一个函数的结果作为输入。这听起来可能很复杂，但是程序员一直都在使用复合函数。下面是Java中复合函数调用的一个示例。
系统，退出。印刷(数学)Pow $(3,2)$);
这是一个复合函数，因为我们取函数pow的结果并将其提供给println。

计算机代写|神经网络代写Neural Networks代考|Understanding Gradients

第一步是计算神经网络的梯度。梯度用于计算特定权重的误差函数的斜率或梯度。权重是两个神经元之间的连接。计算误差函数的梯度可以让训练方法知道它应该增加或减少权重。有许多不同的训练方法都利用了梯度。这些训练方法被称为传播训练。本书将讨论以下传播训练方法:

反向传播

弹性传播

快速传播
本章将重点介绍使用梯度来训练反向传播的神经网络。接下来的几章将介绍其他的传播方法。

什么是渐变

首先，我们来看看梯度是什么。基本上，培训是一种搜索。你正在寻找一组权重，它将使神经网络对训练集具有最小的全局误差。如果我们有无限的计算资源，我们只需尝试所有可能的权重组合，看看哪一个提供绝对最佳的全局误差。

因为我们没有无限的计算资源，我们必须使用某种捷径。从本质上讲，所有的神经网络训练方法都是一种捷径。每种训练方法都是一种聪明的方法，可以找到最优的权重集，而不需要进行不可能的穷举搜索。

考虑一个图表，它显示了神经网络对每个可能权重的全局误差。此图可能类似于图4.1。

计算机代写|机器学习代写Machine Learning代考请认准UprivateTA™. UprivateTA™为您的留学生涯保驾护航。

微观经济学代写

线性代数代写

线性代数是数学的一个分支，涉及线性方程，如：线性图，如：以及它们在向量空间和通过矩阵的表示。线性代数是几乎所有数学领域的核心。

博弈论代写

微积分代写

计量经济学代写

MATLAB代写

2023年5月16日2023年5月16日 CS代写, Neural Networks, 神经网络, 计算机代写

计算机代写|神经网络代写Neural Networks代考|The Error Function

avatest™帮您通过考试

•最快12小时交付

•200+ 英语母语导师

•70分以下全额退款

想知道您作业确定的价格吗? 免费下单以相关学科的专家能了解具体的要求之后在1-3个小时就提出价格。专家的报价比上列的价格能便宜好几倍。

计算机代写|神经网络代写Neural Networks代考|The Error Function

We will start by looking at the local error. The local error comes from the error function. The error function is fed the actual and ideal outputs for a single output neuron. The error function then produces a number that represents the error of that output neuron. Training methods will seek to minimize this error.
This book will cover two error functions. The first is the standard linear error function, which is the most commonly used function. The second is the arctangent error function that is introduced by the Quick Propagation training method. Arctangent error functions and Quick Propagation will be discussed in Chapter 4, “Back Propagation”. This chapter will focus on the standard linear error function. The formula for the linear error function can be seen in Equation 2.1.
Equation 2.1: The Linear Error Function
$$
E=(i-a)
$$
The linear error function is very simple. The error is the difference between the ideal (i) and actual (a) outputs from the neural network. The only requirement of the error function is that it produce an error that you would like to minimize.

For an example of this, consider a neural network output neuron that produced 0.9 when it should have produced 0.8 . The error for this neural network would be the difference between 0.8 and 0.9 , which is -0.1 .

In some cases, you may not provide an ideal output to the neural network and still use supervised training. In this case, you would write an error function that somehow evaluates the output of the neural network for the given input. This evaluation error function would need to assign some sort of a score to the neural network. A higher number would indicate less desirable output, while a lower number would indicate more desirable output. The training process would attempt to minimize this score.

计算机代写|神经网络代写Neural Networks代考|Calculating Global Error

Now that we have found out how to calculate the local error, we will move on to global error. MSE error calculation is the most common, so we will begin with that. You can see the equation that is used to calculate MSE in Equation 2.2 .
Equation 2.2: MSE Error Calculation
$$
\mathrm{MSE}=\frac{1}{n} \sum_{i=1}^n E^2
$$
As you can see, the above equation makes use of the local error (E) that we defined in the last section. Each local error is squared and summed. The resulting sum is then divided by the total number of cases. In this way, the MSE error is similar to a traditional average, except that each local error is squared. The squaring negates the effect of some errors being positive and others being negative. This is because a positive number squared is a positive number, just as a negative number squared is also a positive number. If you are unfamiliar with the summation operator, shown as a capital Greek letter sigma, refer to Chapter I.

The MSE error is typically written as a percentage. The goal is to decrease this error percentage as training progresses. To see how this is used, consider the following program output.

Beginning training…
Iteration #1 Error:51.023786\% Target Error: 1.000000\%
Iteration #2 Error:49.6592918 Target Error: 1.0000008
Iteration #3 Error:43.140471\% Target Error: 1.000000\%
Iteration #4 Error:29.8208918 Target Error: 1.0000008
Iteration #5 Error:29.457086유 Target Error: 1.0000007
Iteration #6 Error:19.4215858 Target Error: 1.0000008
Iteration $\$ 7$ Error:2.160925\% Target Error: 1.0000008
Iteration #8 Error: 0.4321048 Target Error: 1.0000008
Input $=0.0000,0.0000$, Actual $=0.0091$, Ideal $=0.0000$
Input $=1.0000,0.0000$, Actual $=0.9793$, Ideal $=1.0000$
Input $=0.0000,1.0000$, Actual $=0.9472$, I deal $=1.0000$
Input $=1.0000,1.0000$, Actual $=0.0731$, Ideal $=0.0000$
Machine Learning Type: feedforward
Machine Learning Architecture: ?:B->SIGMOID->4:B->SIGMOID->?
Training Method: $1 \mathrm{ma}$
Training Args:
The above shows a program learning the XOR operator. Notice how the MSE error drops in each iteration? Finally, by iteration eight the error is below one percent, and training stops.

神经网络代写

计算机代写|神经网络代写Neural Networks代考|The Error Function

我们将从查看本地错误开始。局部误差来自误差函数。误差函数被输入单个输出神经元的实际和理想输出。然后，误差函数产生一个数字，表示该输出神经元的误差。培训方法将设法使这种错误最小化。
本书将介绍两个错误函数。第一种是标准线性误差函数，这是最常用的函数。第二种是由快速传播训练法引入的反正切误差函数。正切误差函数和快速传播将在第4章“反向传播”中讨论。本章将重点讨论标准线性误差函数。线性误差函数的表达式见式2.1。
式2.1:线性误差函数
$$
E=(i-a)
$$
线性误差函数很简单。误差是神经网络的理想输出(i)和实际输出(a)之间的差值。误差函数的唯一要求是它产生一个你想要最小化的误差。

举个例子，考虑一个神经网络输出神经元，当它应该产生0.8时产生了0.9。这个神经网络的误差是0.8和0.9之间的差，也就是-0.1。

在某些情况下，您可能无法为神经网络提供理想的输出，但仍然使用监督训练。在这种情况下，您将编写一个误差函数，以某种方式评估给定输入的神经网络输出。这个评估误差函数需要给神经网络分配某种分数。数值越高，表示输出越不理想，数值越低，表示输出越理想。训练过程将尝试最小化这个分数。

计算机代写|神经网络代写Neural Networks代考|Calculating Global Error

既然我们已经知道了如何计算局部误差，我们将继续讨论全局误差。MSE误差计算是最常见的，所以我们将从它开始。您可以在公式2.2中看到用于计算MSE的方程。
式2.2:MSE误差计算
$$
\mathrm{MSE}=\frac{1}{n} \sum_{i=1}^n E^2
$$
正如您所看到的，上面的等式利用了我们在上一节中定义的局部误差(E)。对每个局部误差进行平方和求和。然后将结果的总和除以病例总数。这样，除了每个局部误差都是平方之外，MSE误差与传统平均值相似。平方消除了一些误差为正而另一些误差为负的影响。这是因为正数的平方是正数，就像负数的平方也是正数一样。如果您不熟悉求和运算符(大写希腊字母sigma)，请参阅第一章。

MSE误差通常写成百分比。我们的目标是随着训练的进行减少这个错误百分比。要了解这是如何使用的，请考虑以下程序输出。

开始训练…
迭代＃1错误:51.023786％目标错误:1.000000％
迭代＃2错误:49.6592918目标错误:1.0000008
迭代＃3误差:43.140471％目标误差:1.000000％
迭代＃4错误:29.8208918目标错误:1.0000008
迭代＃5错误:29.457086
迭代＃6错误:19.4215858目标错误:1.0000008
迭代$\$ 7$错误:2.160925％目标错误:1.0000008
迭代＃8错误:0.4321048目标错误:1.0000008
输入$=0.0000,0.0000$，实际$=0.0091$，理想$=0.0000$
输入$=1.0000,0.0000$，实际$=0.9793$，理想$=1.0000$
输入$=0.0000,1.0000$，实际$=0.9472$，我处理$=1.0000$
输入$=1.0000,1.0000$，实际$=0.0731$，理想$=0.0000$
机器学习类型:前馈
机器学习架构:?:B->SIGMOID->4:B->SIGMOID->?
培训方式:$1 \mathrm{ma}$
培训地点:
上面展示了一个学习异或运算符的程序。注意到MSE错误在每次迭代中是如何下降的吗?最后，在第8次迭代中，误差低于1％，训练停止。

计算机代写|机器学习代写Machine Learning代考请认准UprivateTA™. UprivateTA™为您的留学生涯保驾护航。

微观经济学代写

线性代数代写

线性代数是数学的一个分支，涉及线性方程，如：线性图，如：以及它们在向量空间和通过矩阵的表示。线性代数是几乎所有数学领域的核心。

博弈论代写

微积分代写

计量经济学代写

MATLAB代写

2023年5月12日2023年5月12日 CS代写, Machine Learning, 机器学习, 计算机代写

计算机代写|机器学习代写Machine Learning代考|Experimentation by solution building: Wasting time for pride’s sake

avatest™帮您通过考试

•最快12小时交付

•200+ 英语母语导师

•70分以下全额退款

想知道您作业确定的价格吗? 免费下单以相关学科的专家能了解具体的要求之后在1-3个小时就提出价格。专家的报价比上列的价格能便宜好几倍。

计算机代写|机器学习代写Machine Learning代考|Experimentation by solution building: Wasting time for pride’s sake

Looking back at the unfortunate scenario of the ML team members building a prototype recommendation engine for the website personalization project, their process of experimentation was troubling, but not only for the business. Without a solid plan in place for what they would be trying and how much time and effort they would be spending on the different solutions they agreed on pursuing, a great deal of time (and code) was unnecessarily thrown away.

Coming out of their initial meeting, they went off on their own as a team, beginning their siloed ideation meeting by brainstorming about which algorithms might best be suited for generating recommendations in an implicit manner. About 300 or so web searches later, they came up with a basic plan of doing a head-to-head comparison of three main approaches: an ALS model, a singular value decomposition (SVD) model, and a deep learning recommendation model. Having an understanding of the features required to meet the minimum requirements for the project, three separate groups began building what they could in a good-natured competition.

The biggest flaw in approaching experimentation in this way is in the sheer scope and size of the waste involved in doing bake-offs like this. Approaching a complex problem by way of a hackathon-like methodology might seem fun to some, not to mention being far easier to manage from a process perspective by the team lead (you’re all on your own-whoever wins, we go with that!), but it’s an incredibly irresponsible way to develop software.

This flawed concept, solution building during experimentation, is juxtaposed with the far more efficient (but, some would argue, less fun) approach of prototype experimentation in figure 3.11. With periodic demos, either internally to the ML team or to the broader external cross-functional team, the project’s experimentation phase can be optimized to have more hands (and minds) focused on getting the project as successful as it can be as fast as possible.

计算机代写|机器学习代写Machine Learning代考|Experimental scoping: Setting expectations and boundaries

We’ve now been through planning of the recommendation engine. We have the details of what is important to the business, we understand what the user expects when interacting with our recommendations, and we have a solid plan for the milestones for our presentations at certain dates throughout the project. Now it’s time for the fun part for most of us ML nerds. It’s time to plan our research.

With an effectively limitless source of information at our fingertips on the topic, and only so much time to do it, we really should be setting guidelines on what we’re going to be testing and how we’re going to go about it. This is where scoping of experimentation comes into play.

The team should, by this time, having had the appropriate discovery sessions with the SME team members, know the critical features that need to get built:

We need a way to de-duplicate our product inventory.
We need to incorporate product-based rules to weight implicit preferences per user.
We need to group recommendations based on product category, brand, and specific page types in order to fulfill different structured elements on the site and app.
We need an algorithm that will generate user-to-item affinities that won’t cost a fortune to run.

After listing out the absolutely critical aspects for the MVP, the team can begin planning the work estimated to be involved in solving each of these four critical tasks. Through setting these expectations and providing boundaries on each of them (for both time and level of implementation complexity), the ML team can provide the one thing that the business is seeking: an expected delivery date and a judgment call on what is or isn’t feasible.

机器学习代写

计算机代写|机器学习代写Machine Learning代考|Experimentation by solution building: Wasting time for pride’s sake

通过构建解决方案进行机器学习实验:为了骄傲而浪费时间

回顾ML团队成员为网站个性化项目构建原型推荐引擎的不幸场景，他们的实验过程很麻烦，但不仅仅是业务。如果没有一个可靠的计划，他们将尝试什么，以及他们将在他们同意追求的不同解决方案上花费多少时间和精力，那么大量的时间(和代码)就会被不必要地浪费掉。

在第一次会议结束后，他们作为一个团队独自出发，开始了他们各自的想法会议，讨论哪种算法最适合以隐式方式生成推荐。在大约300次网络搜索之后，他们提出了一个基本计划，对三种主要方法进行正面比较:ALS模型、奇异值分解(SVD)模型和深度学习推荐模型。了解了满足项目最低要求所需的功能后，三个独立的小组开始在一场善意的竞争中尽其所能地构建。

以这种方式进行实验的最大缺陷在于进行这种烘焙比赛所涉及的浪费的范围和规模。通过类似于黑客马拉松的方法来处理一个复杂的问题对某些人来说可能看起来很有趣，更不用说从团队领导的过程角度来看更容易管理(你们都是自己的-谁赢了，我们就这么做!)，但这是一种令人难以置信的不负责任的软件开发方式。

这个有缺陷的概念，即在实验过程中构建解决方案，与图3.11中更有效(但有些人会认为不那么有趣)的原型实验方法并置。通过定期向机器学习团队内部或更广泛的外部跨职能团队进行演示，可以优化项目的实验阶段，让更多的人(和思想)专注于让项目尽可能快地取得成功。

计算机代写|机器学习代写Machine Learning代考|Experimental scoping: Setting expectations and boundaries

我们现在已经完成了推荐引擎的规划。我们知道什么对业务是重要的细节，我们了解用户在与我们的建议交互时的期望，并且我们在整个项目的特定日期为我们的演示制定了坚实的里程碑计划。现在是我们大多数ML爱好者的有趣部分了。是时候计划我们的研究了。

在这个话题上，我们的指尖上有无限的信息来源，而我们只有这么多时间来做这件事，我们真的应该为我们将要测试的内容和我们将如何去做制定指导方针。这就是实验范围发挥作用的地方。

此时，在与SME团队成员进行了适当的发现会议之后，团队应该知道需要构建的关键特性:

我们需要一种减少产品库存重复的方法。

我们需要结合基于产品的规则来衡量每个用户的隐式偏好。

我们需要根据产品类别、品牌和特定页面类型来进行分组推荐，以实现网站和应用程序上不同的结构化元素。

我们需要一种算法，它将产生用户与物品之间的亲和力，而不会花费一大笔钱来运行。

在列出MVP的绝对关键方面之后，团队可以开始计划解决这四个关键任务所涉及的工作。通过设置这些期望并为每个期望提供边界(时间和实现复杂性级别)，ML团队可以提供业务正在寻求的一件事:预期的交付日期和对什么是可行的或不可行的判断。

计算机代写|机器学习代写Machine Learning代考请认准UprivateTA™. UprivateTA™为您的留学生涯保驾护航。

Category: CS代写

avatest™帮您通过考试

计算机代写|机器学习代写Machine Learning代考|Don’t waste our time: Meeting with cross-functional teams

计算机代写|机器学习代写Machine Learning代考|Experimental update meeting: Do we know what we’re doing here?

计算机代写|机器学习代写Machine Learning代考|Don’t waste our time: Meeting with cross-functional teams

计算机代写|机器学习代写Machine Learning代考|Experimental update meeting: Do we know what we’re doing here?

avatest™帮您通过考试

计算机代写|机器学习代写Machine Learning代考|POST-RESEARCH PHASE DISCUSSION (UPDATE MEETING)

计算机代写|机器学习代写Machine Learning代考|POST-EXPERIMENTATION PHASE (SME/UAT REVIEW)

计算机代写|机器学习代写Machine Learning代考|POST-RESEARCH PHASE DISCUSSION (UPDATE MEETING)

计算机代写|机器学习代写Machine Learning代考|POST-EXPERIMENTATION PHASE (SME/UAT REVIEW)

avatest™帮您通过考试

计算机代写|机器学习代写Machine Learning代考|WHAT DOES THE IDEAL END-STATE LOOK LIKE?

计算机代写|机器学习代写Machine Learning代考|Scope creep: An almost guaranteed assassination of a project

计算机代写|机器学习代写Machine Learning代考|WHAT DOES THE IDEAL END-STATE LOOK LIKE?

计算机代写|机器学习代写Machine Learning代考|Scope creep: An almost guaranteed assassination of a project

avatest™帮您通过考试

计算机代写|数据库代考Database代考|TABLE, ROWS, AND COLUMNS

计算机代写|数据库代考Database代考|RELATIONS, ATTRIBUTES, AND TUPLES

计算机代写|数据库代考Database代考|TABLE, ROWS, AND COLUMNS

计算机代写|数据库代考Database代考|RELATIONS, ATTRIBUTES, AND TUPLES

avatest™帮您通过考试

计算机代写|数据库代考Database代考|PICKING A DATABASE

计算机代写|数据库代考Database代考|RELATIONAL POINTS OF VIEW

计算机代写|数据库代考Database代考|PICKING A DATABASE

计算机代写|数据库代考Database代考|RELATIONAL POINTS OF VIEW

avatest™帮您通过考试

计算机代写|数据库代考Database代考|Retrieval

计算机代写|数据库代考Database代考|Consistency

计算机代写|数据库代考Database代考|Retrieval

计算机代写|数据库代考Database代考|Consistency

avatest™帮您通过考试

计算机代写|神经网络代写Neural Networks代考|RPROP Arguments

计算机代写|神经网络代写Neural Networks代考|Understanding RPROP

计算机代写|神经网络代写Neural Networks代考|RPROP Arguments

计算机代写|神经网络代写Neural Networks代考|Understanding RPROP

avatest™帮您通过考试

计算机代写|神经网络代写Neural Networks代考|Using the Chain Rule

计算机代写|神经网络代写Neural Networks代考|Understanding Gradients

计算机代写|神经网络代写Neural Networks代考|Using the Chain Rule

计算机代写|神经网络代写Neural Networks代考|Understanding Gradients

avatest™帮您通过考试

计算机代写|神经网络代写Neural Networks代考|The Error Function

计算机代写|神经网络代写Neural Networks代考|Calculating Global Error

计算机代写|神经网络代写Neural Networks代考|The Error Function

计算机代写|神经网络代写Neural Networks代考|Calculating Global Error

avatest™帮您通过考试

计算机代写|机器学习代写Machine Learning代考|Experimentation by solution building: Wasting time for pride’s sake

计算机代写|机器学习代写Machine Learning代考|Experimental scoping: Setting expectations and boundaries

计算机代写|机器学习代写Machine Learning代考|Experimentation by solution building: Wasting time for pride’s sake

计算机代写|机器学习代写Machine Learning代考|Experimental scoping: Setting expectations and boundaries

近期文章

近期评论