d = datetime(2018,1,1,[1 2 4 4 2 1 2 3 4 4]',0,0);
t = table(d,rand(size(d)),'VariableNames',["time","data"]);
[C, ia, ic] = unique(t.time,'stable');
d_counts = accumarray(ic,1)
d_counts = d_counts(ic)
t.counts = d_counts;
t_1 = t(t.counts == 1,:)
That said, do you really need to split up the table? Labeling variables numerically like that is usually a sign that you're heading down a path to breakable code. Now that you have the counts in your table, you can dynamically subscript into t and temporarily get the subset that you need. Or, use groupsummary or grouped varfun to apply functions to the whole table, grouped by count.
Best Answer