 All right. So we've already seen how we select other coordinates for both the enrollment and the event as Pamod mentioned before. So you can select coordinates either through enrollment points, event points, or you can have your enrollment, your data element or attribute type of type coordinate. So that also allows you to select the additional coordinates in addition to enrollment and event. As we saw in the track identity instance layer, the track identity layer you are able to plot in the track entities based on their enrollment coordinates. Similar to that you also have event layer which allows you to create clusters based on the event coordinates added in the program stages. So since it takes into account all the events that are created for that respective period and given that the coordinates are available, it will create these event clusters. So if a person has two events in the selected time period, so both the events become part of the cluster because it is taking events into account, not the unique number of patients. So that's the key difference between the two layers. The track identity layer looks at the unique track identity instances in the program. While the events layer will have a look at the events layer will have a look at the total events added in the selected period which may include repeated events for one track identity instance. So that's the key difference between the two. So let's look at one example where we can see the event clusters. So let's have a look at the favorite which is already created. And we can have a look at the layers that have been used here. So you see here you have organization units the boundary layer is at level three. So that has been set accordingly. Then the important is to look at the event layer which is added here for the lab result stage. So if I click on the edit button, I'll see the selections which are made here. So the program. So whenever you want to create a chart with event layer with grouping the events together basis on the basis of their coordinates. You select a program to select the stage and then you select the coordinate field. Now here in this program the event coordinates are the only field available, which is the event location field shown here. But in case if you were collecting coordinates in the program stage in this program stage through a data element also for example additional sets of coordinates. Then you'll have two fields here one for the event location and one for the data element which is also a coordinate type. So you can create clusters on the basis of both the coordinates that you're collecting either the event coordinates or the coordinates which are collected via data element or a program or a or a So both these options are supported. Then you can filter by the event status where they want to see all you want to see active completed so depending upon the status of the event the results will get filtered accordingly. Once you selected the data in the period you can either select relative periods. So in the list of all relative periods are given here. Or if you want to select start in dates then you can select fixed period as well but then the period should be based on two dates start and end date because it will count the events which lie between the dates that you've given. Okay, so for this example we had chosen this year as a relative period. For units, you can in order to keep this chart dynamic, you can select the levels from here. So we want to see all the levels. That is the grand children level at which the all data is collected so you selected two weeks below. So depending upon how your hierarchy is set up you can select the organization units from here. So if you want to keep this chart relative and be flexible enough to meet the requirements for all the users working at different organization units province and districts. For example, then you can select two weeks below because both these kind of users will have access to these are units so they will be able to see the clusters based on their designated districts and provinces. So now we've seen we selected the data we selected the period we selected the organization units. In while creating your clusters, you can put filters and other filters could be of data elements or they could be program attributes as well identity attributes as well. So here's the example you have we have added the data element lab test result. And then we have added the since this had an options that associated to it so we selected that we only want to see the grouping of events where the lab test result came out to be positive. You can select these filters and you can add more than one filter also for example if you're only looking at the mail cases then you can select mail from here. So then it will get revised automatically for now I'll just remove this filter. Then in styling you can style your clusters by different data elements. If you want if your data element has different options associated, then you can select a data element and then the options for that data element will be shown automatically here. Okay, so for example, if I'd selected this then I see all the options associated here, but we want to see the the positive cases by the agenda. So gender has two options male and female so you can choose the color which you want to give to female and which you want to give to male. Okay, so once you make all these selections, you if you want to group these events together then you select group events if you want to see individual events, one event, each event individually then you can select view all events. Okay, if you do grouping then it will automatically create these donut clusters for you why this is being shown as donut because you've done additional styling and have used data element that you want to style by data element which has certain options and then to those options you have assigned two colors. So hence it shows you the result as a donut where you have total three cases so based on the proportion we see here probably two cases are male and one is female so likewise you can design your clusters accordingly. If you wanted to see data for individual events, then you can go to view all events and click on update layer. So you'll see individual events by different colors, but the they will no longer be grouped together. Okay, so these are the individual events which are there. So for each event you can see the gender, the type of test in the lab result which is positive what was the facility, the date time of event or the date of event and the event location. If you want to switch it over back to group events then you can group events and it will again turn into a cluster. So what we'll do now is let us revise this and see from scratch how we can create this one. Okay, so let's try to recreate this one. So I'll go to file, click as new. So I'll start by adding my boundary layer. So I'll go to the boundary layer or the organization units layer, select level three and add layer. So it will automatically add the district boundaries or the boundaries against the level three organization units which are there. Next I'll go and I'll add an event layer. We're into select the variables. So I'll select the program and select lab results. This is the only coordinate field available. So this will automatically be selected. Then I'll go to period. I'll select last one year. Just like this year. Organics, we want to keep it relative and dynamic. So we select the lowest level available. So we select two X below. Now we can add some filters. So from these events, so lab result event will be of all types. You will have lab result event for positive cases, negative cases, inconclusive cases, but you only want to see data for positive cases. So we'll select the data element as lab test result. And since it has options associated. So you, and we only want to see positive cases. So we select as positive. But you can add multiple filters also if you want to. Next you move to style. Now here we want to see donut clusters, which are disaggregated by gender. So we'll select sex as our attribute. And then you have female and male as your options. You can change the color. So you can make the female as pink and male you can select as blue. And you can click on add layer. So by default, you'll always see group event selected. If you want to see individual events, then you can click on the second option as well. Okay, so now I see the clusters are created. So you can do the selections and create your own event layers on the on the on the maps application. So once you have created the the clusters, you can drill down and reach till the individual case. So this is an individual case, and all the data elements which are marked as display in reports in your program stage, they will show they will be shown here in this pop up. Okay, so depending upon what data elements you marked as display and reports and configuring your program stage, you'll have the information for those data elements here. Now, using this layer we are looking at the event data. So this will display all events within that program, and you can filter the events by the data element values, which are there in those events. So if a person was tested twice within the gap of this year, then it will show that person twice because the event happened twice. So this is not unique count of positive cases. These are the count of events where the test result was positive not not for the same person you can have more than one event so you can use events there only when you want to even your focus analysis focus is the number of events for the count of events total events, not unique tracked entity instances. So we have given this requirement to the code developers at Oslo to give us a functionality to also create clusters for unique cases, where we want to see based on say the latest value of the events for example I want to create clusters of ART positive cases who are on treatment. So only want to see those cases who are on treatment. So if I have made three visits in this quarter, then if I do quarterly analysis I should see the patient count unique patient count not the total count of events. So that requirement has come up from any implementation so we already are in the process of sharing the documentation which is required for submitting the feature request so that is under consideration. So once you have created your map with the event layer, you can go to file and click on save. So it's similar to like you do in all analytics model you're saving favorites you can just select positive. Ovid cases, general application and you can save this one. Okay. Then after saving you can use this on the dashboard and you can even use this for your analysis here. Then you can download this map from the maps application. So if you click on download, it will give you some options that if you want to show the name and where do you want to show the legend. So it'll give you four positions. Left, left top corner, left bottom corner, right top corner, right bottom corner. So depending upon what's the most convenient you can select that and it will show the filters that you're filtering cases by lab test result which is positive. So you are styling the results by the sex data element. Okay. So once you've selected your, your legend location you can download this image so you will have this image downloaded on your device. And then you can use this for your presentations. So just to ensure that you capture the entire image so ensure that you have you have the clusters within your image size. So that is a concentration which you need to make that when you're downloading this ensure that you are not in because then it will take up only the area which is shown on the map but not the default size that it shows. So this is how you can create a map on using the event layer. So please one in your learners guide. So please proceed with doing that exercise if there any questions please feel free to unmute and ask a question or put in this chat I'll try to answer that. Now, once I sort of assume the map zoom in the map so I found that the cluster sub in detach and it gives me a values, whereas if I zoom out it so sort of a merge and gives a value. So what is the sort of a fundamental does it have some sort of a sort of methodical way or does it any sort of algorithm or can we sort of make changes to this one. So this is happening based on the coordinates which are there. So when your group when you're zooming in zooming out you're basically drill down on the coordinates so the polygon expands or the cluster expands as you click on the the zoom in feature so then if so basically the coordinates are splitting up into different clusters so if by default if you see you'll have 1259 cases which belong to a group of coordinates which has formed a polygon. If you keep drilling down then the coordinates are keep splitting up so as the coordinates split up the clusters become less and less the in terms of the radius. And if you go come up the coordinates again gets combined. So it kind of forms a cluster of patients who share coordinates within the same parameters which just combines and breaks them together based on the purely on the coordinates. There is no specific algorithm working so this is based on the coordinates which have been selected. So if these coordinates are coming in from one single location or polygon or an area, then it just combines those into one bigger cluster and when you break them it splits into specific coordinates of that location or those events which share that same location in terms of the coordinates selected while data entry is done. Okay, so we can do the exercise one from the learners guide. If there any questions please let us know. If not now then you can put it on the Slack channel we'll answer those. Yesterday you have your grid assignments for event visualizer and the event analytics in the maps. So please don't forget to do the grid exercises and also fill the feedback for today so that your attendance can be mapped. We can check if the feedback is open on modules or if not, we can open that with me.