How to reset count where yyyymm Date gaps exist

[NilR]

How can I add a condition to the below dax to identify the gap between 202209 and 202305 and reset the count?

-- Calc Column
Active_records = 
var _GP = Table[GP]
var indv_id = Table[ID]
var _end_date = Table[DATE]
var _DATE = EOMONTH(_end_date,-24)+1

VAR _COUNT = RANKX ( 
        FILTER (ALL(Table),
            Table[GP] = EARLIER(Table[GP])
            && Table[ID] = EARLIER(Table[ID])
            && Table[MALES_UNDER_35] = EARLIER(Table[MALES_UNDER_35])
            && Table[DATE] >= _DATE
            && Table[DATE] <= _end_date
        ),        
        Table[DATE],
        ,
        ASC,
        Dense
    )
 RETURN 
 
_COUNT

[technolog]

I hope this message finds you well. I've noticed that this solution remain unresolved. If any of you have managed to find a resolution to the issue, I kindly request that you share your solution for the benefit of the entire community. Alternatively, if you're still facing challenges, please do let us know as well.

Your insights and updates will greatly assist others who might be encountering the same challenge.

[NilR]

I could not figure it out!

[technolog]

GP	YYYYMM	DATE	Active_records	Desired Result
33	202105	5/1/21	1	1
33	202106	6/1/21	2	2
33	202107	7/1/21	3	3
33	202108	8/1/21	4	4
33	202109	9/1/21	5	5
33	202110	10/1/21	6	6
33	202111	11/1/21	7	7
33	202112	12/1/21	8	8
33	202201	1/1/22	9	9
33	202202	2/1/22	10	10
33	202203	3/1/22	11	11
33	202204	4/1/22	12	12
33	202205	5/1/22	13	13
33	202206	6/1/22	14	14
33	202207	7/1/22	15	15
33	202208	8/1/22	16	16
33	202209	9/1/22	17	17
33	202305	5/1/23	17	1
33	202306	6/1/23	17	2
33	202307	7/1/23	17	3 Desired Result = RANKX( FILTER( ALL('Table'), 'Table'[GP] = EARLIER('Table'[GP]) && 'Table'[YYYYMM] <= EARLIER('Table'[YYYYMM]) ), 'Table'[YYYYMM], , ASC, Dense )

[technolog]

Desired Result = 
RANKX(
    FILTER(
        ALL('Table'),
        'Table'[GP] = EARLIER('Table'[GP]) && 
        'Table'[YYYYMM] <= EARLIER('Table'[YYYYMM])
    ),
    'Table'[YYYYMM],
    ,
    ASC,
    Dense
)

[NilR · Accepted solution]

I ended up adding new field in sql using lag function then added to my calculated column!

[lbendlin]

What's wrong with using GroupKind.Local in Power Query?

[NilR]

there is more into the actual data and it didn't work!

[Greg_Deckler]

@NilR Seems like a variation on Cthulhu. Cthulhu - Microsoft Fabric Community

[NilR]

I ended up adding flag to my data with sql. Thanks!