rawgraphs / rawgraphs-charts

A curated selection of charts provided through RAWGraphs interface
https://rawgraphs.io/
Apache License 2.0
92 stars 43 forks source link

[Radar chart] adjust axis intervals and enter negative values #112

Closed martidossi closed 1 year ago

martidossi commented 2 years ago

In which part of the interface would this feature applies?

[ ] 1. Load your data [ ] 2. Choose a chart [ ] 3. Mapping [x] 4. Customize [ ] 5. Export

Is your feature request related to a problem? Please describe. The eventual enhancements have to do with the radar chart. 1) I would like to be allowed to set the number of intervals the main axis is divided into. 2) It would be also useful to be able to enter negative values.

Describe the solution you'd like 1) Adding the option in the Customize section with some widget like 'Dots diameter'. 2) Just be allowed.

Describe alternatives you've considered 1) As for the number of intervals, I haven't found an alternative yet. 2) To be able to enter negative values I'm rescaling my data before updating them in RawGraphs.

Additional context /

Thanks.

mikima commented 2 years ago

@martidossi how would you handle negative values? putting the minimum value at the center of the radar? How should we show the 0 value? I think it's a visual issue, if you have suggestions they are welcome!

martidossi commented 2 years ago

Hi, yes exactly, when the minimum value across dimensions is less than 0, I would leave that value at the center of the chart and scale the axes accordingly. Then, if interested in having the 0 value highlighted as a reference, I would perhaps increase the thickness of that circle to differentiate it from the others or change the colour -but ultimately it depends on the problem and it might not always be needed to differentiate the 0 line. What do you think? Anyway, I agree it's more a visual issue, and one could always rescale the data in advance and make them positive; let's say that having the feature implemented would just allow to skip this step and work directly with the original data. Thank you! :)