You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I would propose to fall back to exponent notation ("e" format) for numbers below abs(x)<1e-24 or above abs(x)>1e24.
In my application, usually the numbers are in a range where the "s"-format makes sense. Though, sometimes, e.g. due to numerical inaccuracy of calculations, I get very small numbers, that blow-up my output.
The text was updated successfully, but these errors were encountered:
Yes, true. I currently just check for abs(x)<1e-24 and then use "e" or "s" formatting. My point is more that a 'human' would format 1e-50 rather as 1e-50 than as 0.00000000000000000000000001y and therefore it might be an interesting improvement proposal.
Libraries like vega and vega-lite also use d3-format. There a workaround is also possible with custom formatters, but it gets more and more tedious.
It would be great to improve the 'human-readability' of SI format "s" for very small/large numbers. At the moment d3-format generates:
I would propose to fall back to exponent notation ("e" format) for numbers below
abs(x)<1e-24
or aboveabs(x)>1e24
.In my application, usually the numbers are in a range where the "s"-format makes sense. Though, sometimes, e.g. due to numerical inaccuracy of calculations, I get very small numbers, that blow-up my output.
The text was updated successfully, but these errors were encountered: