Count number of unique values in column one, for each unique value in column 2

I've been faced with a problem that is giving me lots of trouble. I need to calculate the number of unique records in one column for each unique record in the second column. For example: If I have

Activities = {'walk', 'monday'; 'walk', 'monday'; 'eat', 'monday'; 'TV', 'tuesday'; 'run','wednesday'; 'eat', 'wednesday';...}
'call', 'Thursday';'work', 'Thursday'; 'call', 'Thursday';'work', 'Thursday';'sleep','Thursday'}

What I'd like in another column is the total number of unique activies for each unique day, formatted like "Result".

Result = [2;2;2;1;2;2;3;3;3;3;3]

[WAT]

This seems to be a repeat of http://www.mathworks.com/matlabcentral/answers/245839-count-the-number-of-occurrences-of-elements-in-second-column

Answer 1

[Star Strider]

Expanding on Andrei Bobrov’s Answer to a related Question:

[~,~,ic1] = unique(Activities(:,1), 'stable');          % Unique Activities
[~,~,ic2] = unique(Activities(:,2), 'stable');          % Unique Days
H = accumarray([ic1 ic2], 1);                           % Histogram
LH = sum(H > 0);                                        % Sum Logical Array
SH = sum(H);                                            % Sum Event Array
Result = [];
for k1 = 1:length(LH)
    Result = [Result; repmat(LH(k1), SH(k1), 1)];       % Create ‘Result’ Vector
end

[Colby]

Thank you very much. This helps a lot!

[Star Strider]

My pleasure!

To be a bit more descriptive, LH is the sum of different events per day, and SH is the number of total events per day.

[Andrei Bobrov]

[~,~,ic1] = unique(Activities(:,1), 'stable');          
[~,~,ic2] = unique(Activities(:,2), 'stable');          
H = accumarray([ic1 ic2], 1);

Result = repelem(sum(H > 0),sum(H));

or

H1 = sum(H > 0);
Result = H1(ic2);