Forum Discussion
Memory issue, Dax Formula is not being processed
- 10 years ago
Before you move on, you could just try loading less rows or columns and see if that actually fixes the problem.
Regarding unpivot:
Let's say you have the following 7 columns
date, stock, buy, sell, high, low, vol
you can load the table into power query (get data). Then select the first 2 columns, right click and select "unpivot other columns. This will then give you 4 columns as follows
date, stock, attribute, value.
the last 4 column headings will now be text in the attribute column
the values from the last 4 columns will be values in the values column.
Best set thing to do is build a simple test and see how it works.
You ou then need to change your measures.
Instead of
=sum(table[vol)]
you would write
=calculate(sum(table[value]),table[attribute]="vol")
In your biggest table. How many rows, how many columns?
MattAllington in this particular table I have 227,000 rows and 49 columns.
In other table, I have over than 800,000 rows and around 49 columns as well
- MattAllington10 years agoCommunity Champion
49 Colums sounds like the problem. In power pivot, you want long narrow tables. You do not want wide tables where ever you can avoid it.
what is in all these columns? Can you unpivot the data?
- majdkaid2210 years agoHelper V
MattAllington am not quite familiar how unpivot would affect my data. I've never did unpivot exrcise.
Am in FX Trading, and all these columns are related to transactions, exchange prices..ect I've reduced the number of columns to 14, but the DAX formula is still not going through. I won't be able to reduce further as the data would become useless
- MattAllington10 years agoCommunity Champion
Before you move on, you could just try loading less rows or columns and see if that actually fixes the problem.
Regarding unpivot:
Let's say you have the following 7 columns
date, stock, buy, sell, high, low, vol
you can load the table into power query (get data). Then select the first 2 columns, right click and select "unpivot other columns. This will then give you 4 columns as follows
date, stock, attribute, value.
the last 4 column headings will now be text in the attribute column
the values from the last 4 columns will be values in the values column.
Best set thing to do is build a simple test and see how it works.
You ou then need to change your measures.
Instead of
=sum(table[vol)]
you would write
=calculate(sum(table[value]),table[attribute]="vol")