Great tutorial, but I have an issue and I don't understand how to solve it. When I follow all the steps, the lines appear as dots, and not as lines, it remains radial but with dots. Does anyone know why?
Dont know if you are still checking these comments, but I have an issue. For some reason I am getting a single point for each of my dates (years in this case) and not a bar sticking out. Any ideas?
I would need the text to follow the circle outline by following the angles. Doesn't seem possible. If I don't have this option I can't really make the chart I want.
Excellent! I would recommend though, if you are using dates in detail, to multiply the [Radial X] by minus one so that the dates progress in a clockwise direction, which may be more intuitive.
Thanks. The reason we use COS for the X and SIN for the Y is because we are converting the angle to X and Y cartesian co-ordinates. Cos(angle) = Adjacent / Hypotenuse ---> X = Adjacent, whereas Y = Opposite so cos(angle) = X / Radial Normalised Length i.e X = cos(angle) x Radial Normalised Length... the same logic holds for Y...just rearrange.
thank you! this is really helpful!. I have one question, how do we create "Path order" without doing union tables ? because we are using published data sources
Do you know how I can have 24 bins and the length of the bars to be observations within that hour? Basically a bar chart in a circle representing a clock.
can you tell us how to rotate it so january 1st is at 12 o'clock position? On the radial angle formula i just added like 1.7 and that kinda got me there, but is there a more exact way?
I love this tutorial, I have one question tho, when you filter to a certain category, the lines re-align equally around the 360degrees. Is there anyway to filter, and yet have the lines hold their position around the circle, ie in an application of tracking events within an hour?
Hi alienboyx1984, using the current approach you will need to fill the other "hours" with 0 values somehow. once you found the solution, please do share it!
Hi, I'm having a problem with my radial bar chart. I'm trying to plot different fossil species from four different localities. I wanna show the percentage of fossils per locality so there should be four categories however I cannot seem to plot them in bar charts; they end up as points in the graph.
Hi Iris. It might be a good idea to ensure that the formulas are exactly the same as the ones used in the tutorial, as even a slight difference may cause an incorrect value which will distort the graph. Also, to ensure that you select a Line chart as chart type and include a Path variable in the file you are connecting to.
I used this demo to create my own Radial Chart. It works great! Except for when I add a filter to the view. The whole graph shifts to different axis measures. How can I fix it?
Hi, I have a problem. It gives me an error here for the Radial Normalized Length: [Radial inner] + IIF(ATTR([Calculation1]) = 0, 0 SUM([Radial field])/WINDOW_MAX(sum([Radial field])) * ([Radial outer]-[Radial inner]))
Hi Lorenzo, Please check the formula matches as below: [Radial Inner] + IIF(ATTR([Path Order]) = 0, 0, sum([Radial Field])/WINDOW_MAX(sum([Radial Field])) * ([Radial Outer]-[Radial Inner])) You can also download the workbook from the site below: www.superdatascience.com/yt-tableau-custom-charts-series/
Really useful tutorial. Thank you ! Question: How can I adjust this Radial Graph to look like a "Radial Lollipop Chart". I want to add a sizable dot on the outer edges of the bars.
Hi Herman, Thanks for your inquiry. You can combine our Lollipop chart tutorial below with you radial chart to crate the desired effect: ua-cam.com/video/l9G5lCL4obE/v-deo.html
Good Tutorial, I tried to implement it even though i did not understand the formulas... the reason why they are included, some how managed to create it though.. can someone please give me the explanation for each formula over there. TIA
Hi, thanks for your video! I did the graph with your dataset and it worked perfectly. However, then I tried with other data and I have a problem. Everything is fine until I choose the dimension to filter by colours. Then the circle disappears and I can only see three lines. I tried everything, even change the names of the cities from your dataset and it does not work anymore. I would really appreciate if you could help me. Thank you.
Hi Paula. It is difficult to tell not having insight into what your data looks like, but my best guess will be that it might have to do with the data prep. Ensure that you union the dataset onto itself and create the "Path Order" variable. It is also important to be aware of what is being included in the the level of detail. We have used the date field (continuous day) and by including the city is also taken into account into the level of detail.
Hi, It finally worked. I am gonna explain the solution here in case anyone else needs it. In the example the City field is geographical data. In my case, I had names of cars which is just text data. It did not work because I was having the names of cars as Dimension, just like in the example, but it should be Attribute (ATTR). Then it worked. Thanks :)
Hi Paula, Your explanation is very helpful--I had the same issue, and in my case, I had a column called "project names" instead of cars. But now, when I changed "project names" to an Attribute, it got rid of the color differentiation completely. The function changed all of the project names to *, do you have any suggestions to fix?
Vivon Crawford Hi! It is not easy to say... according to my experience it can only work with one. It has to be either a dimension or an attribute. It should directly work with a dimension. Check your data type is text and not anything else. I am sorry I cannot be very helpful but if it doesn't work either as dimension or as attribute maybe it is something else :/
It gives me an error here for the Radial Normalized Length: [Radial inner] + IIF(ATTR([Calculation1]) = 0, 0 SUM([Radial field])/WINDOW_MAX(sum([Radial field])) * ([Radial outer]-[Radial inner]))
Hi Monika. You will notice that for each of the different cities (colours), the line lengths are correctly proportioned - but not when all are combined. Using the method described in the tutorial, the line lengths are determined per City. If you remove City from the level of detail from your visualisation, you will see that the lines are drawn correctly proportioned taking all data into account.
Ok i was able to do the entire tut, but this seem look nice if you have a lot of rays/line. since the detail he used was the date each ray/line represent each day but if use like a category, i would only have a few line which makes a lot of spaces in between. i tried increasing the size of the line but it doesnt look good. thank you though. i hope you can also create tut something like these ua-cam.com/video/R1TO7Pf0erY/v-deo.html
![Lp. Сердце Вселенной #60 РОЖДЕНИЕ ЛОЛОЛОШКИ [Финал] • Майнкрафт](http://i.ytimg.com/vi/YoR0pAV9FVQ/mqdefault.jpg)
The best video I have seen for creating Starburst Chart in Tableau! 💯
Thanks for your comment!
Great tutorial, but I have an issue and I don't understand how to solve it. When I follow all the steps, the lines appear as dots, and not as lines, it remains radial but with dots. Does anyone know why?
Thank you! How can I draw the circle without using on date field ?
Dont know if you are still checking these comments, but I have an issue. For some reason I am getting a single point for each of my dates (years in this case) and not a bar sticking out. Any ideas?
I would need the text to follow the circle outline by following the angles. Doesn't seem possible. If I don't have this option I can't really make the chart I want.
Excellent! I would recommend though, if you are using dates in detail, to multiply the [Radial X] by minus one so that the dates progress in a clockwise direction, which may be more intuitive.
Thanks for your suggestion bctoledo, it is helpful indeed :)
Excellent, I was about to ask the question!
Can someone please share the bike image, without background? Please!
HI ! Will these always be the formulas to use for all radial charts?? Is this a standard?
Thank you!
Works! great tutorial!
Thanks. The reason we use COS for the X and SIN for the Y is because we are converting the angle to X and Y cartesian co-ordinates. Cos(angle) = Adjacent / Hypotenuse ---> X = Adjacent, whereas Y = Opposite so cos(angle) = X / Radial Normalised Length i.e X = cos(angle) x Radial Normalised Length... the same logic holds for Y...just rearrange.
thank you! this is really helpful!. I have one question, how do we create "Path order" without doing union tables ? because we are using published data sources
thank you! this is really helpful!. I have one question, how do you add the trip (km) label to the graph?
Do you know how I can have 24 bins and the length of the bars to be observations within that hour? Basically a bar chart in a circle representing a clock.
Nicely done. Concise.
How to achieve this without duplicating data by union?
Great video! I totally gonna use that one! thanks!
so blurry screen did you film this under water?
Thanks for this great video, do you know if I can sort the line based on their values?
Nice tutorial
I just have one question.
how you link elements on the map, so that they can respond to the radial chart?
best,
Hi Francisco, you need to adjust Dashboard Actions as shown in the tutorial
Nice! One question, do you know how to do a union on a extract from server?
Hi Nikola,
Please check this thread - community.tableau.com/ideas/7076
can you tell us how to rotate it so january 1st is at 12 o'clock position? On the radial angle formula i just added like 1.7 and that kinda got me there, but is there a more exact way?
Hi Tron,
I'm glad you are experimenting with this visual. Your solution works well for this project.
Hi Tron you can an float parameter as an adjuster
Can you share the dataset ?
Videos are good but it would be nice if you could explain the calculated field formulae as well.
I love this tutorial, I have one question tho, when you filter to a certain category, the lines re-align equally around the 360degrees. Is there anyway to filter, and yet have the lines hold their position around the circle, ie in an application of tracking events within an hour?
Hi alienboyx1984, using the current approach you will need to fill the other "hours" with 0 values somehow. once you found the solution, please do share it!
Hi, I'm having a problem with my radial bar chart. I'm trying to plot different fossil species from four different localities. I wanna show the percentage of fossils per locality so there should be four categories however I cannot seem to plot them in bar charts; they end up as points in the graph.
Hi Iris. It might be a good idea to ensure that the formulas are exactly the same as the ones used in the tutorial, as even a slight difference may cause an incorrect value which will distort the graph. Also, to ensure that you select a Line chart as chart type and include a Path variable in the file you are connecting to.
I used this demo to create my own Radial Chart. It works great! Except for when I add a filter to the view. The whole graph shifts to different axis measures. How can I fix it?
Hi Hernan,
Thanks for adding to this thread.
Could you clarify the type of filter you are adding to this visual?
Thank you!
Hi, I have a problem. It gives me an error here for the Radial Normalized Length:
[Radial inner] + IIF(ATTR([Calculation1]) = 0, 0 SUM([Radial field])/WINDOW_MAX(sum([Radial field])) * ([Radial outer]-[Radial inner]))
Hi Lorenzo,
Please check the formula matches as below:
[Radial Inner] + IIF(ATTR([Path Order]) = 0, 0, sum([Radial Field])/WINDOW_MAX(sum([Radial Field])) * ([Radial Outer]-[Radial Inner]))
You can also download the workbook from the site below:
www.superdatascience.com/yt-tableau-custom-charts-series/
Really useful tutorial. Thank you ! Question: How can I adjust this Radial Graph to look like a "Radial Lollipop Chart". I want to add a sizable dot on the outer edges of the bars.
Hi Herman,
Thanks for your inquiry.
You can combine our Lollipop chart tutorial below with you radial chart to crate the desired effect:
ua-cam.com/video/l9G5lCL4obE/v-deo.html
Thank you SuperDataScience very helpful
It would be helpful if you included the formulas you copy and paste so that we can do that as well.
Calculations:
Radial angle
(INDEX() - 1) * (1/WINDOW_COUNT(COUNT([Radial Field]))) * 2 * PI()
Radial Field
Trip duration
Radial Normalized Length
[Radial Inner] + IIF(ATTR([Path Order]) = 0,0, sum([Radial Field])/WINDOW_MAX(sum([Radial Field])) * ([Radial Outer] - [Radial Inner]))
Radial X
[Radial Normalized Length] * COS([Radial Angle])
Radial Y
[Radial Normalized Length] * SIN([Radial Angle])
Parameters:
Radial Inner
0.5
Radial Outer
1
Great tutorial
Good Tutorial, I tried to implement it even though i did not understand the formulas... the reason why they are included, some how managed to create it though.. can someone please give me the explanation for each formula over there. TIA
me too, wish there were more explanation on the why
Hi, thanks for your video! I did the graph with your dataset and it worked perfectly. However, then I tried with other data and I have a problem. Everything is fine until I choose the dimension to filter by colours. Then the circle disappears and I can only see three lines. I tried everything, even change the names of the cities from your dataset and it does not work anymore.
I would really appreciate if you could help me.
Thank you.
Hi Paula. It is difficult to tell not having insight into what your data looks like, but my best guess will be that it might have to do with the data prep. Ensure that you union the dataset onto itself and create the "Path Order" variable. It is also important to be aware of what is being included in the the level of detail. We have used the date field (continuous day) and by including the city is also taken into account into the level of detail.
Hi,
It finally worked. I am gonna explain the solution here in case anyone else needs it.
In the example the City field is geographical data. In my case, I had names of cars which is just text data. It did not work because I was having the names of cars as Dimension, just like in the example, but it should be Attribute (ATTR). Then it worked.
Thanks :)
Awesome, glad to know that you found a way to make it work!
Hi Paula,
Your explanation is very helpful--I had the same issue, and in my case, I had a column called "project names" instead of cars. But now, when I changed "project names" to an Attribute, it got rid of the color differentiation completely. The function changed all of the project names to *, do you have any suggestions to fix?
Vivon Crawford Hi! It is not easy to say... according to my experience it can only work with one. It has to be either a dimension or an attribute. It should directly work with a dimension. Check your data type is text and not anything else. I am sorry I cannot be very helpful but if it doesn't work either as dimension or as attribute maybe it is something else :/
Amazing job
Anyone that’s experienced in Tableau knows the hardest part of any viz is removing those border lines. It’s always a shot in the dark. 😂
Really Amazing！！
Thanks for your feedback Alex!
it says unknown function IFF called while setting the RADIAL_NORMALIZED_LENGTH
Hi Nikhil. It is, in fact, the IIF function, not the IFF function.
It gives me an error here for the Radial Normalized Length:
[Radial inner] + IIF(ATTR([Calculation1]) = 0, 0 SUM([Radial field])/WINDOW_MAX(sum([Radial field])) * ([Radial outer]-[Radial inner]))
guys, the length of the line doesn't correspond to the actual Trip Distance. There're almost same size bars with distinct values.
Hi Monika. You will notice that for each of the different cities (colours), the line lengths are correctly proportioned - but not when all are combined. Using the method described in the tutorial, the line lengths are determined per City. If you remove City from the level of detail from your visualisation, you will see that the lines are drawn correctly proportioned taking all data into account.
Hey, exactly, good point :)
😎 Magic! 🙌🏼
Ok i was able to do the entire tut, but this seem look nice if you have a lot of rays/line. since the detail he used was the date each ray/line represent each day but if use like a category, i would only have a few line which makes a lot of spaces in between. i tried increasing the size of the line but it doesnt look good. thank you though. i hope you can also create tut something like these ua-cam.com/video/R1TO7Pf0erY/v-deo.html
Hi Chadd, we will check that out
wohoo.... now im excited...
Lukka!
Awesome :D