Here is our training plan.
Our agenda for this training module is to cover brief OI@OWS Ecosystem, the D.A.R design pattern, the qlikview core technology, introduce qlikview desktop, and the qlikview architecture. And finally, to cover the OI application architecture.
OWS platform consists three layers: PaaS
(Amazon AWS), Micro services layer, an APP layer. We re-engineer AWS to create Huawei PaaS, OWS Plus. OWS+ can support Web App, Mobile App, System Integration and workflow orchestration.
The second layer is micro-service layer. It has various domains. Huawei has opened this layer to encourage 3rd parties to create micro-services of specific area.
The top layer is App layer. This is built on top of the micro-service layer. Each App is the combination of multiple micro-services. Our three flagship Apps are SDM, OI and WFM. AABD(automatic alarm behavior discovery) is algorithm centric.
The main functions of OI@OWS include two aspects: one is data presentation, the other is report. both of them support customization.
the information of network element, SDM@OWS, WFM@OWS system, the work order, employee , fee can be customized to form a management Dashboard， to support the daily management needs of CEO, CTO, CIO and operations and maintenance supervisors. In addition to supporting display on a computer screen, Dashboard also supports demonstrations on mobile phones and PADs to facilitate managers'daily reward needs.
At the same time, for the operation and maintenance analysis expert team, OI supports multi-dimensional data analysis, can support experts to carry out root cause analysis from multi-dimensional and multi-way, locate problems, and make continuous improvement possible.
This is the D.A.R pattern to manage the business. D means Dashboard, A means analysis, R means report. There are some advantages of this pattern as below:
照着ppt念
放视频
Then next is two short video to demonstrate the usage of oi.
after that, I will let the customer use these demo to practice. And I will answer their question and guide them to use it.
OK, next. Let’s take a look at the qlikview core technology. There are several advantages to using qlikview. Firstly, queries and qlikview are significantly faster than traditional relational databases and OLAP Data wharehouses, because all data is loaded into ram. Computers are able to access ram faster than they can access the hard disk, all solid state drive. Secondly, when the user makes any selections in filters or charts, qlikview will instantly recalculate the entire data model and update all relevant visualization. Thirdly, qlikview is capable of easily combining data from multiple data sources. We will examine this concept further in advance when we create mashups of custom data and OI data. Qlikview is capable of accessing data in many different data sources. We'll take a look at the possible data sources later in this session. There are many other advantages to using qlikview. I will talk about it later.
Qlikview Desktop is a tool used for developing qlikview applications. Qlikview Desktop is a standalone windows application that allows users to write their own ETL scripts, create data models, and design visualization. There are over one hundred fifty different functions available for users to transform their data. Common transformations include cleaning, aggregation and logical processing. Qlikview assists users in creating their own visualization through a set of embedded /im baded/ components that allow users to quickly and easily create charts, tables, and KPI figures to display their data.
The physical qlikview file is qvw,The qvw is created in qlikview desktop and contains all the information needed for a dashboard, or report ,the load script, data model ,data and user interface are all stored within the q v w .qlikview is capable of using a wide variety of data sources, including ,o d b c and OLDB for connecting to standard databases. Qlikview can also load data directly from excel files and c s v files. There also exists a wide range of extra connectors available for connecting to more specialized data sources, including sap and salesforce.
qlikview makes use of its own proprietary data store called qvd ,qvds are created in qlikview desktop, but rather than saved as you would a normal file. They are saved using a function in the load script, each qvd stores a single data table and can only be used by other qlikview applications. The diagram on the right displays a typical qlikview deployment using both qvw s and qvds .A special type of QVW called a qvd generator is used to extract the data from the data source and store as qvds. This QVW only contains a load script and does not contain any visualization or store any data. The QVD created by the QVD generator is then used as a data source for a suite of qlikview reports. Byprocessing your data through QVD generators, there are several advantages. Firstly, these QVD generators function as the ETL layer of qlikview and become a single source of truth. Using QVDs also increases development speed and flexibility, as users do not need to write lengthy load scripts repeatedly. Finally, and most importantly, QVWs load data from qvds significantly faster than other data sources. What takes minutes to load from a different data source can take seconds using qvds.
The standard qlikview architecture is comprised of three distinct sections, one for creating content, one for reloading, publishing, and distributing content, and one for consuming content. All content is created using the qlikview desktop. This is typically done on the business uses machines. And when the user is ready to share their qlikview applications, it is transferred from their machine to the qlikview server, the qlikview server is made up of three distinct components, the qlikview server itself is used to load qlikview applications, user session information into ram. The qlikview server also uses the cpu to re calculate the data model when selections are made.The qlikview publisher will run the load scripts of all configured applications located on the qlikview server.The qlikview publisher can also function as a back in security tool that is not covered in these training sessions. The qlikview access point is a central location, the way users can access all qlikview applications stored on the server. Users are able to consume content from the qlikview server through the use of various qlikview clients. The primary clients you will be using are qliview desktop and the ajax client through your web browser. Although another common client used is the internet explorer Plug-in. Access to the qlikview application can be controlled through the use of various security methods such as active directory, windows file security, or qlikview security. 这个对于本次课程不是很重要，所以简单了解一下就可以
Here is the oi application architecture version three. Once again, starting at the bottom, we have the data sources. These are the primary data sources from which all data used by oI originates. The data here is then transformed by the etl layer and stored in the OI data warehouse. The oi data warehouse functions as a single source of truth for all benchmarking and project inside reporting. qlikview is the business intelligence tool used for all benchmarking and project inside reporting. OI supports users adding their own data through the use of the custom data upload. All files that are uploaded are stored in the OI data warehouse as physical files that can then be used as a data source for qlikview reporting.
This is the foundation knowledge two module. Titled setting up Qlikview. This module is the last in the foundation knowledge modules before moving into the develop module.
The agenda for today's training session is to go with the installation process for qlikview personal edition review how and why you should be licencing a license. And finally, we'll take a look at the qlikview desktop.
16
具体的安装路径
I knew in advance that your operation system are all windows 7 and 64 bit.So I prepared the installation package for your operation system.Ok,let start.Double clik the exe.
First of all, select the language that you like to install. And there's a wide variety of languages that you can use, including chinese. But i'll be using the english installation language. It. Ok, the installation process is also very fast. It's like next except the user license agreement. Next, again, you can enter in your name, if you like, the next. Generally, the default destination folder is the best one to insulate into, say, next again, you want to use the complete installation. So hit next again, and then install once the installation is finished.
Select finish, and it will exit the installer. Now that you have installed Qlikview.
17-22.
After that, you will need to lease a license. The version of Qlikview that you have installed is called the qlikview personal edition. This is the version of qlikview that is free for everyone to use and develop. There are, however, a few restrictions with the personal edition, firstly you cannot open up qvws that other people have created. Secondly, other users cannot open up qvws that you have created.In order to upgrade your Qlikview from the personal addition to the full QLIKVIEW desktop, you will need to lease a license from the server. Doing so will give you full access to qlikview desktop, and you'll be able to open up any qvws that you like.
In order to lease a license from the server, you will first need to open up qlikview desktop. You can very easily tell that you're in the personal edition of qlikview as up on the top left. You can see qlikview x sixty four personal edition to upgrade your personal addition to the full version, click on file open url and entering the url that you see on your screen. Now, the user credentials you need to enter in this sign in page. I'm not your standard use credentials, but rather generic user credentials for leasing licenses.
They'll username and password are now being displayed on your screen into those when you've entered the access point. The final step for leasing your license is to click on the application that you see here called golf quest, though qvw once the application has loaded, you can verify that Qlikview has been licensed by looking at the top left. You'll see that the personal edition has vanished. Thank you, has not been licensed. Now the Qlikview has been licensed.
23-25.
I'm just going to give you a quick overview of some of the functions of qlikview.
First of all, let's close this application and this access point and create our own application. We do that by using the blank paper icon at the top. And you here we have the getting started wizard. I recommend that you don't use this. And in fact, that you disable it coming up. You do so by d selecting this check box here. It's okay. And then the cross to close. And here we have the blank canvas for developing qlikview.
Along the top. You have a number of functions. You can create another new document. You can open an existing document, or you can save your current document. You also have the edit script button and the reload button, along with the various chart created visits. Let's take a close look at the script editor here. We can see the reload button and debug button, save and print. Here on the main is where you'll be entering in most of your load scripts, and then down the bottom, we have the data loader system.
Now we will be using the data loader to help us until advanced developer session one. But it's a good idea to get an understanding of what you can do with it. For example, the various database connections we have available are here. We can connect to those databases. And we can have a look at the method, which is in them by using the select button. Here, we can load table file, such as excel files and c s v . We can load from other qlikview files using binary load.
Or we can learn data from web files. in order to show you some of the functionality of qlikview, i'm going to load up some dummy data. You can do that by pressing ctrl q q and here we have simple data set. In order to load this data into qlikview, you have to click the reload button. But for that, you need to save it as a file. You do that by pressing. Ok, i'm just going to save here in the default location as new.
Okay,So the data is now being reloaded, and it's loaded into memory. Here's a list of the available fields that you can use. We have several dimensions, dimensions one, two, and three, and several expressions, expression one, two, and three. We're going to select dimension one and hit add. Now, what we're doing here is we're taking the dimensions which we have available, and we're adding it to the list of dimensions that we want showed in this boxes. We hit. Okay, we now have a list box for dimension one.
Now, one of the core concepts in Qlikview is the idea of green, white, and grey. So green is when you selected something. So we're gonna select a and you can see that it's gone green, because it's being selected. now, if we bring in a second field, dimension two, you'll see that there are several values in dimension two, which are white, a and b what this means is that a and b on dimension two have data which is related to the capital a in dimension one.
The gray values mean that those values have no relationship to the selection which has been made for dimension one. So green, white, grey, green is what you selected. White is where there is a relationship. And gray is when there's no relationship to your current selections. So i've shown you the script editor. Now we're gonna take a quick look at the table view on up the top. Right? As you can see here we have a visualization of the various tables of data that we floated, the fields that are common between tables, I indicated by a line between them, in this case, we have an unfilled as the key-linking the three tables together.
You can gain a better understanding of what's in the table by right clicking and selecting preview. And this will show you a list of all the data there. We're gonna close this, okay, this is the sample introduction of qlikview.
26
The developer one training module-Developing analysis chart in Qlikview. In this module, we will cover how to create some simple visualization in qlikview.
Our agenda for this training module is to cover some simple visualization, including the bar chart, line chart. Pi chance stacked and grouped, bar, chart, pivot and straight table and k p I. we will take a look at when to use these visualization, some advantages and disadvantages of each. Then we will cover how to create these visualization in qlikview.
然后使用Chart Introduction Application 演示
The developer one training module-Developing analysis chart in Qlikview. In this module, we will cover how to create some simple visualization in qlikview. Our agenda for this training module is to cover some simple visualization, including the bar chart, line chart. Pi chance stacked and grouped, bar, chart, pivot and straight table and k p I. we will take a look at when to use these visualization, some advantages and disadvantages of each. Then we will cover how to create these visualization in qlikview.
Here we have a qlikview application displaying all in mentioned visualization. This application is available in the material I shared before. And I recommend you follow along yourself when I start creating the chance later, starting from the top left, we have the bar chart, line chart. Pie chart. Stacked bar chart. pivot table, straight table and kpI .these are just a few of the available visualization available and qlikview. And although there are more complicated visualization available, we recommend that you keep your designs as simple as possible for, as einstein said, everything you should be kept as simple as possible, but no simpler.
With so many visualization available to use, it can often be tough to decide which one is the most appropriate. Here is a chart that will help you with making that decision. Ask yourself, what would you like to show? And then followed the decision tree to the chart, for example, if I would like to show a comparison, among items. One variable per item, few categories, and few items. I end at the column shot. There is a wide range of visualization in this chart and qlikview is capable of showing most of them.
Now we will move on to a more detailed explanation of each of the charts on a simple visualization page, starting with the bar chart. The bar chart is best used when you wish to display one dimension and one expression. The data to be displayed should be normal. That means it is in distinct categories such as countries, people and products. The reason that we recommend you use nominal categories. It's because the white space between bars implies its distinction between the values. In the example we have on the left, we have resellers being displayed by count of orders.
The resellers are distinct category, and hence a good fit for the bar chart, as you can see with the bar chart, it is very easy to compare bar size. And we can improve the information clarity by placing the expression value at the end of the bar. The unfortunate consequence of bar charts, clarity of information, is that they often overused and the user will become bored of them.
The next child we will look at is the line chart. The line chart is best used when you wish to display one dimension and one or more expressions or you wish to use one or two dimensions and one expression.
The data to be displayed should be ordinal. That is, the data follows a specific order. The best example of ordinal data is time. The reason we recommend using ordinal values is because of the nature of the line indicates some relationship between the value as opposed to the gaps between the bars in the bar chart. As mentioned, the lines a great fishing, a relationship between the values on the chart and hence a perfect for displaying time. Another great advantage to line charts is due to their small size. You can run multiple lines on the same shot very easily.But like the bar chart, it is very easy to overuse line chart.
Next up is the pie chart. The pie chart is a very controversial chart with many visualization experts preferring never to use them. Instead favoring the bar chart in every single use. That said, there are a few special use cases where you'd want to use a pie chart instead of a bar chart. Like the bar chart, a pie chart is ideal when you have one dimension and one expression. But with one caveat, due to the area constraints of a pie chart, you shouldn't use them when there are more than six values, as the individual slices of the pie will become too small to easily distinguish. The main advantage and reason you would want to use a pie chart over a bar chart is because it is very easy to compare the size of each individual slice to the overall pie.
So when your goal is for the user to compare individual slices against the total pie, then the pie chart is best. Pie charts are also an appealing alternative to bar charts when you already have too many bar charts. In general, users will find it harder to compare the area of a slice than they would comparing the size of a bar. And if you wish to convey exact values, you will need to either attach a legend with the values to the chart, or have the values on the chart itself, which may look on tidy.
The next chart is the stacked and grouped bar chart. The stacked or grouped bar chart should be used when you wish to display multiple related dimensions and one expression, or you wish to display one dimension and multiple related expressions. Like the bar chart, the data to be displayed should be nominal. For example, countries, people and products. When deciding whether to use a stacked or group chart, you should consider which data is more important to display. If it is more important to compare the value of the primary dimension, then you should use stacked. As you can see on the chart on the top left, the stacked bar chart, it is easier for the user to compare the overall bar size to one another than it is to compare the sizes of the individual bars within the greater bar.
Alternatively, if it is more important to compare the value of these secondary dimension, then you should use a grouped bar chart. If we look at the group's bar chart down the bottom left, as you can see, it is easier for the users to compare the size of each individual bar. But comparing the sum of both bars together, it's much harder.
The pivot table is the first of the two tables that we will be looking at. The pivot table is best used with multiple dimensions and expressions.
There should be some relationships between the dimensions as it is possible to drill into a pivot table or create cross tables. Pivot tables are good when you need to see exact values, they can be salted in various ways to highlight different dimensions. And our good way of displaying large amounts of data. Additionally, cross tables are a great way to slice one dimension by a second one. Like in the example on the left, where we are slicing YEAR by Quarter. The trade off, is that it's not as easy or quick to use a table as it is a chart, also with cross stables, it is best to use a secondary dimension with many unique values, otherwise across table will be too large. The straight table is the other type of table available in qlikview. The main difference between this table and the pivot table is that you cannot drill into this table, nor can you make a cross table. Once again, like many of the simpler chance, a straight table maybe overused.
The final visualization object we will be looking at is the kpi. k p i's are best used when you just want to display one piece of information such as account or some. The main advantage over k p I is that it allows a user to gain insight into a single piece of information quickly and easily. The trade off, is that it shows no comparisons or relationships.
Now that we have discussed when you should use each visualization as well as covered off some of the advantages and disadvantages of each them. Well now take a look at how you create them. Before you start creating your charts. It's a good idea to have an understanding of the data. You can gain an understanding of the data by having a look at the table viewer, that can be selected at the top, on the table view icon. Here, you can get an insight into the data model.
And if you wish to gain a greater understanding of the data in the each table, you can right click on the table and select preview. This will then show you a preview of the data. Once you're done previewing the data, you can close this by pressing. Ok,
in this qlikview application, I have given instructions to have to create each of these visualization under their respective tabs. In these tabs, you can see the steps taken to create the perfect visualization. I am not going to show you how to create each of these visualization.
But the steps that I take maybe different than the steps outlined here. The first shot that will be creating is a bar chart. to create a visualization and qlikview first, right click anywhere on the work space. This will then open up the right click menu, move to a new sheet object, and then down to chart. this will open the chart wizard. The chart wizard is what you will use to create most of your visualization in qlikview. in the chart wizard, You can choose from any of the available chart types, ranging from bar chart to line chart pie chart, and then some of the more advanced visualization, such as the radar chart, the grid chart, and the block chart. For this demonstration, I will be creating a bar chart. So I was selected bar chart and press Next, this opens up the dimensions page. Here,You can choose any dimension that you wish to create a visualization from. In this case, I wish to use contract ame. So I will type in contractor name, move my mouse onto it and click add.
This will add contract a name to the list of use dimensions on the right. Now that I have selected the dimension that I wish to use, I can click next. This will bring up the expression editor where I can enter my expression. like excel functions, such as sum and count can be used here. And for the purposes of this demonstration, I will be using count. qlikview supports intellisense. So if I start typing the field that I want ,qlikview will find it for me. in this case, I want to count the orders.
So the field that I will count is orderid, as you can see, the intellisense has finished the orderid for me. Ok, here is a completed expression at this point, I can now click finish and qlikview will create the job for me. Here we have the basic chart, but there are quite a few things that we can do to make it look a bit nicer. in order to edit this chart simply, right click on it and go to properties, with the chart over here, on the left, on the chart properties over here on the right, I will be able to make changes to the chart and you can see them as I make them. the first change that I would like to make some of the dimensions tab. I go up the dimensions selected. And here I want to remove the label, this dimensional label here, because the contract name is inherently implied by the names here, it is not required for me to have the contract a name label. You can remove it by d selecting the label check box. I will now hit apply so you can see what this does, as you can see, contract a name label has vanished.
Next up, there are no values here on the chart. Null values may be useful in some cases to show where data is missing. But in this case, I do not wish to show them. You can hide no values by selecting the suppressed when value is null checkbox here, as you can see, the null values have disappeared. The next thing I do is add a title to my chart. I can do that on the general tab. under show title in chart,that is my title.
Hit apply! You can see that that is now being carried across the chart here. Next, I would like to sort my chart. I can do that on the sort tab. In this case, I wish to sort my chart by the y value. to do that, I move across to sort by select y value descending and hit apply. Now a chart is being sorted by the y value descending. next, we'll make some changes in the expression tab. I wish to share the values and the data point.
I do that down here under the display options and select the values on data points Check box. Now you can see that we have the values on the ends of the bars, show the customization. I wish to change the axes, here we can change the primary dimension labels on the dimension axis. We will make them slanted. Now you can see that the primary dimension labels have become slanted. A little bit easier to read. We can also change the font here. by selecting font, We'll leave them at the same size and font type, but changed it to Bold.
It ok, and then apply. And I can see that the bolt is carried across here. If you wish to change the font of the expression values, you can do that underneath the expressions front. But for the cases of this example, we won't be changing that. The chart is looking better, but there's still more improvement. This number compact to be a little bit difficult to read. So we will put a comma to indicate the thousands, to do that, move across the number stab and select the fixed two decimals and hit apply.
Then we have a common value. It is possible to choose many different number formats. You can choose numbers to scientific precision decimals like we have money. And just to show you what that looks like, we'll select it and hit apply. And as you can see, he's put dollar sign in front and put two decimal places. You can also show dates, times, and intervals. You can also show the numbers as percentages. Let's move it back to fix two zero decimals. A chart is almost complete, but there's a few things left we can do.
If we move across to the layout tab, we will change our border color to black, and it will remove rounded corners. Next We remove the caption, they have It, looks pretty good. If you were to create a chart like this, i'm sure any client would be happy. But there are a few more customizations we can do. I would consider these customizations optional. So let's go back to bar chart,right click on it, Properties, now, because we have the values at the ends of the bars, we actually don't really need to display the values here anymore, so we can hide them.
If we move cross to the axis time and click hide axis. As you can see, the axis is now managed, the advantage of hiding the axis is that it gives you a little bit more real estate for you to customize your chart. So if you have a very full-on dashboard, like a popular here in huawei, I recommend that you hide the axis. Next up. We can change the color of the chart. Oh, from the color step here, you can see a list of the colors available.
The color that's being used is the first color in the one box, but we select multi colored. We get a different color for each bar. Usually, I don't recommend that you use multi coloured bars, as it may confuse the users in this case. So we'll turn off multi colored and hit apply. We can change the color of all of the bars. We do that by clicking on the color, clicking on the base color fixed. I'm in choosing the color that we wish to use, in this case, we'll select this nice probably maroon color, okay, is it okay?
Then hit apply. Another color is carried through fuller bars. This is now a very nice Huawei looking bar chart. But what if we want to change it a little bit more? Well we can change it from a vertical bar chart to a horizontal bar chart. And we do that from the style tab under orientation. So if we select the sideways orientation, hit apply, you can see that the possibly oriented horizontally. I recommend that you use this particular orientation when you don't have very much vertical space.
See what I mean? I can change the size of this chart by clicking on the edges where you have this two arrows and dragging which way I want to go. In this case, I want to reduce the vertical size. like that. And now you can see, even though the visualization has much less vertical space, so quite easy to use. Whereas if we convert it to vertical bonds, doesn't look quite as nice. So we're gonna convert it back horizontal, and leave it like that.
The next shot that we'll look at is the stacked or group bar chart. So we're gonna take our current thought shot, and we're going to copy it. We can copy charts in qlikview in a number of ways. You can simply right click to copied to clipboard. But there's an easier way to do that. First of all, find the edge of your chart. See here where we have the four arrows. So if you click and hold here, you can drag your thought around, but I don't want to drag my chart around.
I want to make a copy of it. To make a copy of my child. I follow the same steps. I find the edge, I click and I hold. But then I also hold down ctrl on my keyboard. You see how there's a little plus underneath the four arrows that indicates that the child will be copied. I now let go of my mouse and we have a copy of the chart. with this chart copy, We can now make some changes. Let's make it a little bit bigger by clicking and dragging at the bottom like this.
And now we can add a second dimension to this chart by right clicking on it, go to properties, go to dimensions. Now under the fields here, we want to choose a second dimension to cut the first one by, in this case, we will add the date_flag, I find date_flag by typing in date. It's what for me. Then I click add, and it'll automatically be added to the use dimensions. If we hit apply, we'll see what this child starts to look like, as you can see here, we're now cut the contract the name by complete flag and create flag.
This shots already looking pretty good, but there are some more customization. So we can do to make it a little bit easier to use. For starters, let's remove that maroon color and restore to default colors. We do that by going to the colours tab and selecting get default colors, as you can see, that maroon color has turned into a blue color. You could apply, that change we carry through. That's it. Ok. So as you can see here, legend over on the right is taking up a lot of space.
We can change that by moving where the legend is. to move the legend, make sure you've selected the chart by clicking on it. Then hold ctrl shift. This brings up the advanced chart menu. And you can move around each parts of the chart. In this case, i'm going to move the legend to the bottom. I do that by clicking and dragging down to the bottom. Now, pay attention here, you'll see that this small rectangle will become a large rectangle.
When that happens, you know, you've reached the bottom, let go of the mouse, and your legend will be moved to the bottom. You can move your legend anyway you want by clicking and dragging around. You can also move your title around and change the size of your visualization .Sometimes we know moving the parts of the chart around,it can get a bit buggy. For example, you happen to lose the legend. You can restore it by right. Clicking on the chart go into properties, select the general tab, and select reset user sizing and reset user docking, then hit Okay. as you can see, the chance has been restored to its default settings.
Let's return the legend to the bottom of the chart. Yeah, we have a completed grouped bar chart. What if we want to convert it to a stacked bar chart,How can we do that? Well, if we right click on the chart, go to properties, style tab. Then you have two sub types, grouped and stacked, let's convert it to stacked. And here we have a stacked bar chart. You can also change the orientation here, much like you would a regular bar chart. There is one last thing for us to do with stacked bar chart and that's to name it. So if we move across the general tab, change the name here to stacked bar chart, and then hit Okay. And then you have stacked bar chart title on a stacked bar chart.
The next chart that we want to use, it's a line chart. So to create the line chart, we right click on the work space. New sheet object,and chart,Then we select line chart here. And we'll call it line chart. Then hit next. Next up.
The dimension that we want to use is a little bit different this time. We want to use one of the time based dimensions. Dimention that I want to use is reportyearmonth, we select reportyearmonth ,add, then hit next. The expression we use the count of orderid, okay, finish. And here you have your very basic line chart. But we want to apply some customization to it. We do that by right click properties. We go back to the dimensions time.
Once again, we're going to hide the label and suppress when value is null, already starting to look better Now. as you can see along the bottom, there's quite a lot of dates coming up, now we're gonna address this the same way that we do with bar chart by going to access and changing the primary dimension labels to slanted. There we go. Now we can see more of ah, date labels along the bottom. Next up. Let's change our number format. Once again, we do that in the numbers tab, changed two fixed zero decimals and apply. personally, I'm a fan of using a curved line, rather the straight line.
And we can do that under the expressions tab. line. Change that to smooth. That's looking nicer. We could set values on data points. But if we do that, you'll see this quite a lot of overlap. And it can be difficult to use, especially in situations where you have two points which are right next to each other. It can be very difficult for the user to read them. So I recommend that you hide values and eight points. But as a trade off, one axes, we can show the grid. It apply. You see now that you have this grid line moving across at each of the intervals. This makes it a little bit easier to use. Once again, we're going to go across to the lay on top. We're gonna change our borders to black. And we're going to remove the rounded corners. This is looking good. Let's remove the caption, apply. This is looking very nice, to show you a little bit more functionality in a line shot. I'm going to add a second expression, just as important, count of orderid divided by two.
Now we have a second expression in line chart. To help the user distinguish between them. We're going to give them labels. If we select our first expression, we can enter an expression one. we can enter expression 2 to as a label here. Next, we hit apply. Those changes will carry through. We also want to add a reference line. in order to add a reference line. We do under the presentation tap, go to reference lines ,add,here. You can enter in the values for the reference line.
I'm going to enter in a simple expression of ten thousand. I'm gonna change the color of my line, tomorrow and up the weight three, hit apply. And you see that you have a reference line added there. A child is almost complete, but we just need to move the legend. I'm just dragging the the legend down to the bottom like that. There we have it, a completed line shot. The next shot that we're going to create is the pie chart.
We create the pie chart much like we did the other chance, right click on the work space, new sheet object, chart, from here, we select pie chart on the chart type. I'm going to call it ,pie chart. Then it next ,select dimensions and type in the dimension that you'd like to use. In this case, we're going to use contract the name again, hit add, next. So the expression count orderid, then hit okay, and finish, you know, have a simple pie chart available for use.
Now that we have our basic pie chart, there's quite a few things that we can do to edit it, to make it a little bit more usable. in order to do that, right Click on the chart,go to properties. So the first thing that we're going to do is we're going to change the style. There's quite a few styles available under look. I prefer simple look, but there's a lot that you can choose from. I definitely do not recommend that you use a three d as it can be difficult for the user to understand.
So we're going to select the simple one of the top left. hit applied. And they have a much nicer looking pie. Next up, we're going to move the null values from dimensions. Then we're going to change the legend a little bit. If you go to presentation, we can show the numbers in the legend. When you're using a pie chart. I highly recommend that you showed the numbers in the legend, because without the numbers, it's very difficult for the user to understand the exact numbers, free slice of the pie.
So we select that checkbooks, hit. Ok, now gotta numbers also sorted by y value descending, much like we did with our other charts. It's looking pretty good. So there are some simple things that we can do to make it look even better. You can change the number format. We can change the color of our borders to black, removed around corners. hit apply, looking much better. It's also removed the caption. Ok, now we have a nice pie. It's very easy to use them to understand.
We were only using five different values. So there's only five slices of the pie, which is less than six. This is all looking good. So there's still a few more things that we can do to make a pie look a little bit better. First things first. Let's move across to the expressions tab and select values on data points. We now have the value from the data points. I personally don't recommend that you use this version. I prefer that you put the numbers on the legend, as you can see, we've got a little bit of overlap here, looks a little bit ugly.
Also by putting the values on data points, it's made the pie significantly smaller. To serve point of comparison, i'll remove it. And now you can see that we have the big pie. So if you leave your pie chart like this, then this is a nice looking pie. And it's very easy to use. The next visualization that we'd like to use is a table. So i'm just gonna move this pie chart out of the way, just on here. Now to make a table, we do it the same way that we've created all these other visualisation right click new sheet object chart.
And then you have two options for your tables. Here. You pivot table and a straight table. I'm gonna start with a pivot table. as you can see, the show title in chart is gray. The reason this is gray is because table doesn't have a title, as you'll see in a moment, we hit next, we can choose our dimension as normal. I'm going to use my favorite contract the name stop by typing in selected, add next for an expression.
ok and then finished. As you can see here, we have a very simple table that we can use. Now, there's more customization that we can do to make a little bit more usable, right Click on the chart and hit properties. So what we'll do is we're going to remove uh, when value is null like that, we're also going to change the expression label as you can see that the expression label is quite obvious, is a little bit ugly. So we can change this now to what account hit upon.
And then you can see that change is carried through. Now we're going to change the borders to black ,removed around corners. But we're gonna leave the caption there, because otherwise, it's very difficult to tell what this table is showing. in this case, We'll leave it as well. Account. apply. Now we have a nice simple table, female changes that we can do to make a little bit nicer. We can sort by y value descending. No, I have a nice sort. Let's have a second dimension to this table.
We do that by going to dimensions type and finding a second dimension that we want to add. In this case, we used a flag again, injured in hit apply. And now you can see we have little plus sign. This is because what we're using is a pivot table. We can expand out this table by hitting the pluses. We can undo that selection by hitting clean. Yeah. So we can expand out each of these like that by hitting the pluses. So here we have our finish product for pivot table.
But there's a couple of customization is that we can do to make it a little bit nicer looking. First things first is that we can convert this hoover table into across table. Now we do that by selecting the column that we want to move, now i'm clicking and dragging right now. And you can see this blue arrow. That blue arrow indicates where this column is going to end up. So if I look over here, it will move. But if I click and drag across the left, we can actually get to swap places with the other field.
And now we have date flag first. But what we want to do is we actually want to put date flag across the top here. So we do that by clicking and dragging, hold it, and then you see where it gets to their the arrows up the top. Right? We let go. We're now sliced contract the name by date. Flag, can also make this column a little bit easier to read by dragging it out, much like you wouldn't excel. And now you can quite clearly see that that says contract a name and a flag.
Now I just realized that the number format doesn't have comments. So let's have comments. Do that by right click on the job properties, the format again, fixed zero decimals hit apply. There we have it. So this is a nice looking table. You can leave it like that and be very happy with, but we want to change it. So we're gonna copy it. Now. We're gonna copy this chart, just like we did the bar chart earlier. In order to do that, you select anywhere on the caption, you hold down control, you click and drag. When you got the plus sign, let go of your mouse.
And there you go. You've got your coffee with this copy of the pivot table. We're going to create a straight table. We do that by right clicking on the table go into properties moving across the general tab, and then selecting straight table. If we get applied, you see that it's immediately be swept over to a straight table. But the swap has done some funny things to our table for status is added in a little total row at the top there. And it's also changed the ordering of our dimensions.
So let's move across the sort up. Now, the difference with a straight table is that cones a given priority for sorting. In this case, we wanna order count to me the top priority. We do that by selecting it and hitting promote until this at the top, as you can see here. Now, old account is at the top. We want to change it by numeric value descending. This means that will have the largest values at the top. That's looking much nicer. Next, we want to change the total row.
So that's at the bottom. And we do that on the presentation tap and the turtles here. We have the option to have on the first row or the last row. And i'm gonna move it to the last row. I'm also going to get the label. Do that by hitting the use label checkbooks. Then enter the label that you want it apply. You'll see here now that we have total label and the total row. This is looking good. We might just leave it at that.
So now we have uh, pivot table and or straight table next to each other. As you can see, the two different ways of showing the same data. Some users may prefer to use this cross table format, whilst other users may prefer just to use a straight table format. The final visualization that will be creating is a k p I so k p i's are created completely differently to how you create charts. The object that we use to create a k p I from is actually a text object in order to create your text object, you're right, click on the work space, new sheet object.
And then you move down to text object I was gonna hit. Okay, so you can see what a basic text object looks like. As you can see, it has this blue colour with a rounded borders. Let's customize it a little bit. Why click properties and the text you can enter and the value that you want. Now, text box is capable of showing simple strings. If you just type in your string like this, you see this text books, but you can also enter and expressions instead by starting with count.
Now, this process is a lot like excel. So we're gonna try and count of already and hit apply, as you can see. Now we have the expression being displayed. Now, changing the format of the number in a text box is a little bit different than changing on the chart. As you can see, there is no number tab. So in order to change the number former, we have to use the numb function. So the non function takes into number. It's the first parameter. And then the format that you want to display it as, that's a second parameter, in this case, we'll be using this format here.
It apply. And as you can see, now a number has a nice format to it. So there's some more customization is that we can do. The first thing i'm going to do is i'm going to change the color to white looking good. I'm going to make the font larger and bold. We use twenty. Both makes it a little bit easier to read. Move across to the layout tab. We're going to use borders. We're gonna make them black. We're going to move the rounded corners.
Ok. This is looking very good. So if you have to leave it like this, it had looked pretty nice. What does it mean? There's no labeling. The user can never understand what we're looking at. So we're going to add a label to it. The way in which we do that is we add a second text object over the top of this object, and then we put the label left. How we do that? Well, first of all, we're gonna make this a little bit smaller. Then you right click on it, go to properties again, select right? Horizontal alignment, bottom vertical alignment.
Now it's down the bottom, right? We're also going to change the layout layout. It's a custom level. Three can apply then. Okay, now we're going to copy it and paste the copied directly on top of it. We do that by selecting it. You mean ctrl c then control v and now we have our second textbooks on top of this one. We can double check that by dragging the top one along over there. And as you can see, we now have two separate text boxes.
If we want to restore this one where it was, we just undo it by hitting ctrl z now we want to make some change. It's a tough one. So when I click on that and go properties to ensure that this one stays at the top, we want to put it on customer will for to land. Next, we move across the general tab, an internal label. What account we need to change the background transparency to one hundred percent so that we can see through it and see both the old account and this value here.
But we also need to change the layout of our text so that they're not sitting on top of each other. So if we change it to the top right and had a point, you can now see that we have all account of the top left and the value being displayed at the bottom. Right? There are few will change that we can do to make it a little bit easier to read book. We can change the font of the top layer will make it a little bit smaller so that people are more drawn to the the number instead of the label that's looking good just as one last change.
I'm going to select both text boxes by dragging around them. Now have both been selected. I'm gonna make it a little bit smaller. Yeah, this is looking good. So if you to leave, you can't be like this. You have a k pi. That's very easy to use. I recommend that you leave your k p I like this because it's very simple and very easy to use. The last visualization that we're gonna look at today is the quickly selection boxes. So these are created mean much the same way that you've created all the other visualisation by going new sheet object list box.
You then choose the field that you'd like to use. We're going to use our good old friend, contract the name slightly and then hit. Okay. And here you have your very simple selection box. I recommend that you keep your selection boxes as simple as possible, because they fit in very nicely with the overall click theme of green, white, and grey, as discussed in the early video. But what you can do if you want to make a look a little bit nice. So whilst keeping the simple theme is you can change the border, the blank and hit apply can also remove the rounded corners.
And then we have alice box, which fits in very nicely with our overall thing. This is the end of the developer one module. And in the developer to module will be covering the o I data model.