可视化仪表板是什么_可视化库优化开发了丰富的仪表板

可视化仪表板是什么

As you are going through the last analytic mile in your project to visualize and expose powerful insights to Line of Business users, you will undoubtedly need to leverage visualization libraries to display consumable and rich metrics, allowing your users to make the right decision, at the right time, within context.

在您进行项目的最后一个分析里程以可视化并将强大的见解呈现给业务部门用户时，毫无疑问，您将需要利用可视化库来显示消耗性和丰富的指标，从而使您的用户可以在在适当的时候，在上下文中。

As it is often said, a picture is worth a thousand words. This holds true in the analytics world. A chart is worth hours of data analysis done in advance for Line of Business users by data scientists, data engineers or power users. Picking the right visualization is just as important as any other task when it comes to dashboard creation. Developers have to pick which visualization library works best for the experience they are trying to build within their dashboards.

经常说，一张图片值一千个字。 在分析世界中，这是正确的。 对于业务线用户，数据科学家，数据工程师或高级用户预先完成的图表价值数小时的数据分析。 在创建仪表盘时，选择正确的可视化与任何其他任务一样重要。 开发人员必须选择哪种可视化库最适合他们试图在其仪表板中构建的体验。

并非所有可视化库的构建都相同 (Not all visualization libraries are built the same)

Two things are true in the current visualization world: visualization libraries are ever growing and, as with most software solutions, there is no silver bullet. You could be looking at a low level solution like D3 or a ready to use one like NVD3. You have libraries with a few ready to use visualizations like Morris.js and you have Chart.js with a richer selection at the expense of higher configuration costs. Each of these libraries behave very differently. Committing to one of them usually implies a high cost if you need to change them down the road. This may pose an issue when you figure out you need a specific visualization your library doesn’t offer.

在当前的可视化世界中，有两件事是正确的：可视化库在不断增长，并且像大多数软件解决方案一样，没有万灵丹 您可能正在寻找像D3这样的低级解决方案，或者像NVD3这样的现成可用的解决方案。 您拥有一些可以立即使用的可视化库，例如Morris.js ，而Chart.js的选择范围更广 ，但代价是配置成本更高。 每个库的行为都大不相同。 如果您需要在将来进行更改，那么承诺使用其中之一通常会意味着高昂的成本。 当您确定需要图书馆不提供的特定可视化效果时，这可能会引起问题。

I-See visualization 1

我看可视化1

In our previous blog titled “Last mile of dashboard development”, we introduced the challenges of dashboard development. IBM Emerging Technologies is working on prototyping and incubating a dashboard development framework code named “I-See”. This framework, targeted at developers, is already used to drive some IBM internal Proofs of Concepts (PoCs) and is getting richer every sprint. Among many other features to speed up time to value, I-See offers developers a way to mix and match different visualization libraries into a single unit that can be plugged into the dashboard rendering flow. It removes the necessity to commit to a single library at any given point.

在我们之前的博客“ 仪表板开发的最后一英里 ”中，我们介绍了仪表板开发的挑战。 IBM Emerging Technologies正在致力于原型和孵化名为“ I-See”的仪表板开发框架代码。 针对开发人员的这个框架已经用于驱动某些IBM内部概念验证（PoC），并且每个冲刺都变得越来越丰富。 I-See在众多其他功能中可以加快实现价值的速度，它为开发人员提供了一种方法，可以将不同的可视化库混合并匹配到一个单元中，该单元可以插入仪表板渲染流程中。 它消除了在任何给定时间点提交到单个库的必要性。

渲染流程 (Rendering flow)

One of I-See’s main tasks is to create visualizations. It does by following key simple steps:

I-See的主要任务之一是创建可视化。 通过执行以下关键的简单步骤即可：

1. Fetch data from an endpoint. The concept of endpoints in I-See is very simple. They provide a very easy way for developers and administrators to expose simple or complex data services to end users to consume via meaningful endpoints.
2. Transform data into an I-See supported format. Since not all data is provided in the same way, I-See goes through a transformation pipeline that enables developers to take control of the process. We’ll get into a more details in a future blog post.
3. Invoke a renderer and pass transformed data and configuration.

1. 从端点获取数据。 I-See中端点的概念非常简单。 它们为开发人员和管理员提供了一种非常简单的方法，可将简单或复杂的数据服务公开给最终用户以通过有意义的端点使用。
2. 将数据转换为I-See支持的格式。 由于并非以相同的方式提供所有数据，因此I-See经历了一个转换管道，使开发人员能够控制该过程。 我们将在以后的博客文章中详细介绍。
3. 调用渲染器并传递转换后的数据和配置。

Each exposes a pillar in I-See’s process: data fetching, data conversion and data visualization. Each one gets abstracted under different elements within I-See. The one that deals with visualization is called a Renderer.

每一个都暴露了I-See流程的一个Struts：数据获取，数据转换和数据可视化。 每个人都在I-See中的不同元素下抽象。 处理可视化的一个称为渲染器。

渲染器：可视化解释器 (Renderers: the visualization interpreters)

I-See visualization 2

I-See可视化2

In I-See, a Renderer is a special element in charge of communicating between I-See’s rendering flow and the actual visualization library. Its main job is to transform the configurations that are I-See specific into ones that can be understood by the target library. By introducing these elements, I-See is able to communicate with all sorts of visualization libraries. It comes bundled with a few renderers like D3, NVD3, C3 and Morris.js. As a developer, you are not constrained to using any of those. By creating your own renderer, you teach I-See how to communicate with your favorite visualization library.

在I-See中，Renderer是一个特殊元素，负责在I-See的渲染流程和实际的可视化库之间进行通信。 它的主要工作是将特定于I-See的配置转换为目标库可以理解的配置。 通过引入这些元素，I-See能够与各种可视化库进行通信。 它捆绑了一些渲染器，例如D3 ， NVD3 ， C3和Morris.js 。 作为开发人员，您不受任何限制。 通过创建自己的渲染器，您可以教I-See如何与您喜欢的可视化库进行通信。

I-See renderers can be mixed and matched to create an even more powerful one, giving developers a wider range of possibilities for visualizing data. This is especially useful when a specific visualization is not provided by the main library, but by another one. If I-See renderers exists for those two libraries, then you can fully or selectively combine them to obtain a new renderer that knows how to create the visualization you are looking for, using a different library for it.

I-See渲染器可以混合和匹配以创建功能更强大的渲染器，从而为开发人员提供了更广泛的可视化数据可能性。 如果主库未提供特定的可视化，而另一库未提供特定的可视化，则此功能特别有用。 如果存在这两个库的I-See渲染器，则可以完全或有选择地将它们组合以获得一个新的渲染器，该渲染器知道如何使用其他库创建所需的可视化。

移动仪表板示例 (A mobile dashboard example)

Performance tracking

绩效追踪

As we are socializing the I-See framework with early adopters, we created a Proof of Concept (PoC) for a sophisticated but efficient mobile dashboard for an organization training high performance athletes in Canada.

当我们与早期采用者进行I-See框架的社交活动时，我们创建了一个概念证明（PoC），用于复杂但高效的移动仪表板，用于为加拿大的高水平运动员提供培训的组织。

This dashboard exposes real-time, live data from athletes captured during practice and provides coaches with immediate feedback, enabling timely corrective actions. We leveraged the D3 library, customized the visuals to the sport to capture data as athletes are practicing.

该仪表板可显示练习期间捕获的运动员的实时实时数据，并为教练提供即时反馈，从而能够及时采取纠正措施。 我们利用D3库，为运动定制了视觉效果，以在运动员练习时捕获数据。

This dashboard will evolve overtime, capturing more valuable data – bio-mechanical, physiological, … – via Watson IoT and a set of properly targeted sensors.

该仪表板将随着时间的推移而发展，并通过Watson IoT和一组正确定位的传感器捕获更多有价值的数据-生物力学，生理数据等。

Note: organization and athlete names were anonymized to maintain data privacy.

注意：匿名的组织和运动员姓名是为了维护数据隐私。

下一步 (Next steps)

翻译自: https://www.pybloggers.com/2016/11/visualization-libraries-optimization-developing-rich-dashboards/

可视化仪表板是什么