Group by

[uday0994]

Below is the Example data.

Table1:

a,

b,

c,

d,

e,

f,

g,

h,

i,

j,

k,

l,

m

Table2:

Load

a,

b,

c,

sum(d) as d,

avg(e) as e,

avg(f) as f,

count(distinct (g)) as g

resident Table1

groupby  a,b,c;

drop Table  Table1;

Now, I want direct field g also into the dashboard without aggregation. so when i add g into table2 and write g in group by, the d,e,f values are changing due to newly added field in group by. when i take g into seperate new table and left join with Table2, it is working for 1000 lines ,but when we go for full reload, it keeps on loading and after 4-6 hours it is failing.   I have tried applymap also. Could anyone please explain me on how to achieve this.

[Gysbert_Wassenaar]

Sounds like there are multiple g values per a-b-c combination then. So which of these multiple values of g should be used? FirstValue(g), LastValue(g), Maxstring(g), Minstring(g) ? Perhaps you can give us an example of what you're trying to achieve.

[uday0994]

g is a dimension. I need g field directly without applying any aggregation and also the field which we have used count(distinct(g))

[uday0994]

in the above aggregations only average and count of g values are changing in the front end when we write g in groupby. The sum agg values are working fine.