How to get the number of entries of histograms in the legend?

Question

I plot three histograms (originating from a large Dataset ds) in one graph

  Histogram[{#"Var1" & /@ Normal[ds[Select[#"Var2" == 1 &]]],
  #"Var1" & /@ Normal[ds[Select[#"Var2" == 2 &]]],
  #"Var1" & /@ Normal[ds[Select[#"Var2" == Null &]]]},
  {0, 50, 2} , "Frame" -> True, 
  "FrameLabel" -> {"performance", "counts"}, 
  "ChartLegends" -> {"XY: test 1", "XY: test 2", "XY: unknown" },
  "ChartStyle" -> {Red, Green, Blue}, "PlotLabel" -> "Test"]

and want to display the number of entries per histogram in the legend (where the XY can be seen).

But I didn't find a way to get the number of entries per histogram. Is there a function existing?

Answer 1

You can use BinCounts for that purpose, e.g. in that fashion:

Block[{bw = {0, 50, 2},
  ds1 = #"Var1" & /@ Normal[ds[Select[#"Var2" == 1 &]]], 
  ds2 = #"Var1" & /@ Normal[ds[Select[#"Var2" == 2 &]]], 
  ds3 = #"Var1" & /@ Normal[ds[Select[#"Var2" == Null &]]],
  bc1, bc2, bc3},
 {bc1, bc2, bc3} = (ToString@Total@BinCounts[#, bw]) & /@ {ds1, ds2, ds3};
 Histogram[{ds1, ds2, ds3}, bw, "Frame" -> True, 
  "FrameLabel" -> {"performance", "counts"}, 
  "ChartLegends" -> {bc1<>": test 1", bc2<>": test 2", bc3<>": unknown"},
  "ChartStyle" -> {Red, Green, Blue}, 
  "PlotLabel" -> "Test"]
 ]

Answer 2

I'm not exactly sure what you want but here's the solution to a possible interpretation.

Here are some test data

ds = RandomVariate[NormalDistribution[1, 1], {10^4}];

The histogram of which is

h = Histogram[ds];
(* output not shown here *)

Now there's a function HistogramList[] which gives

hlst = HistogramList[ds]

(* Out[324]= {{-(13/5), -(12/5), -(11/5), -2, -(9/5), -(8/5), -(7/5), -(6/5), -1, -(4/
   5), -(3/5), -(2/5), -(1/5), 0, 1/5, 2/5, 3/5, 4/5, 1, 6/5, 7/5, 8/5, 9/5, 
  2, 11/5, 12/5, 13/5, 14/5, 3, 16/5, 17/5, 18/5, 19/5, 4, 21/5, 22/5, 23/5, 
  24/5, 5}, {3, 3, 11, 17, 21, 42, 59, 81, 121, 195, 264, 338, 426, 558, 639, 
  701, 736, 810, 787, 694, 719, 658, 567, 419, 325, 252, 184, 126, 90, 64, 45,
   21, 12, 4, 3, 0, 2, 3}} *)

from which the number of entries in the histogram is easyly extracted:

Length[hlst[[2]]]

(* Out[325]= 38 *)