Hi,

I have been working on this to make it clearer. On the powerpivot tab 'accuracy' you will see the measures that relate to the excel page 'MAPE by Area'. I have done stage one as you advised building 'Mapemeasure' to measure accuracy and then Gradearea average which brings back straight averages as we discussed.

I have now tried to apply weighting, I want to apply weighting to everything above grade in the row labels so that is family, vendor and area. In accuracy measure you will see my attempt to do this and the weightings but it has failed - I hope this helps you understand my issue more, let me know if not and I will see what I can do.

https://onedrive.live.com/redir?resid=3AE8882B3E11F58!119&authkey=!ABqtkoPSKxaYXTo&ithint=file%2cxlsx

Many thanks

Hello,

I have a power pivot data model on Sharepoint. After I enabled the scheduled refresh which works fine, I noticed something strange. When I open the workbook in excel it gives me error that the model might be damaged and it couldn't be loaded. I have ebabled refresh on open and on Sharepoint works without failure, but locally doesn't and this is troublesome because I can't edit the data model. Is there something I can do?

Thank you!

Andreas

Hello Experts,

We dynamically create .ODC files for Excel pivot table connections to SSAS databases. Those are published in SharePoint and the library path is configured in client profiles through a registry key set at logon. That way users can browse existing connections with friendly names without having to remember technical details when initiating a connection.

We'd like to do the same with the existing connections in PowerPivot (stand alone in Professionnal edition) so users can access relationnal data in a self-service orientation, but it doesn't seem to work the same way.

The connections physical locations don't seem to be the same as they are directly in Excel. The network connections option is not available, and the local connections don't seem to point to the MyDataSources folder / Office Queries folder as in Excel. The ideal solution would be to be able to deploy .ODC files in a shared location and simply add the path to this location through registry manipulation at logon.

Any ideas ?

Thanks !

This is what I do.

Method:

1. I make an export of a Business Objects query to an Excel file, save it to my local drive. Our BO can only eport to Excel in xls format (65.000 rows)

2. In Excel, part of my Office 365 ProPlus account, I create a Power Pivot sheet. I import the Excel data from step 1 into the power pivot workbook (get external data, from other sources, then I select the saved Excel file)

3. I import some other excel files that mostly contain segmentations I want to add to the original BO export

4. I create some pivot tables on a (data) worksheet, still all in the same excel workbook. The pivot table contains the selections and data I want to visualize.

5. On other sheets in the same file I create graphs (Bar charts, Pie charts etc) that call on the pivot tables from step 4.

6. I Save and close my file.

Purpose:

The above steps allow me to rerun the queries in BO anytime I want, save the data in exactly the same manner as described in step 1, refresh the Power Pivot sheet and all my graphs will automatically present the new data.

Problem:

When I reopen the Excel Workbook with the Power Pivot data and the graphs Excel this error message appears:

"We found a problem in some content in "My file name.xlsx. Do you want us tot try and recover as much as we can? If you trust the source of this workbook click Yes."

When the file opens after recovery:

- the power pivot sheet no longer exists 

- the pivot tables on my data sheet are gone, the data is still there, but only as values

- the sheets containing the graphs that call on the pivot tables are also gone. What I would expect is for the graph to be empty because the pivot table disappeared. I would not expect that the sheets themselves would disappear. They are physically removed from the file.

My QUESTION:

Am I doing something wrong, is the method I use not allowed? If so, which step should I perform in a different way?

Or is there something else wrong?  

Any help is appreciated.

Kind regards,

Erik van Gurp

Dear All

I have received common error when creating relation:

The relationship cannot be created because each column contains duplicate values. Select at least one column that contains only unique values.

The issue is strange because I am sure that there are no duplicate values in my table! I checked it.

What I've done was to create a single column "Append column" by append few columns with articles from different databases using Power Query. After that I removed duplicates.

The final column with unique values was added to powerpivot model as a new table. There are no duplicates and spaces.

Then I tried to create the relation between the column with append values and other databases which I had used to create this column.

So it is look like this:

Details of error:

Error Message:
============================

The relationship cannot be created because each column contains duplicate values. Select at least one column that contains only unique values.

============================
Call Stack:
============================

   at Microsoft.AnalysisServices.Common.RelationshipController.CreateRelationship(DataModelingColumn sourceColumn, DataModelingColumn relatedColumn)
   at Microsoft.AnalysisServices.Common.RelationshipController.CreateRelationship(String fkTableName, String fkColumnName, String pkTableName, String pkColumnName)
   at Microsoft.AnalysisServices.Common.SandboxEditor.erDiagram_CreateRelationshipCallBack(Object sender, ERDiagramCreateRelationshipEventArgs e)
   at Microsoft.AnalysisServices.Common.ERDiagram.ERDiagramActionCreateRelationship.Do(IDiagramActionInstance actionInstance)

============================

What could be  a reason of this issue?

I am creating some analytics content for clients in Excel 2013 using PowerPivot.  The data itself is being exported to CSV files and I am importing the data in Power Pivot and adding it to the data model. Note that the data can get very large, on the order of millions or rows.  

This is all well and good, but I do not want to have to recreate the whole workbook every time I export my data.  Furthermore, I do not want to have to recreate everything for every single client. I would like to create a "template" of this setup, and be able to swap in the data on the fly "automagically", without having to manually update the connection in Excel.  Is this possible?  

Hi all,

I've been trying to get a semi-complex measure out for over a day now but seem to be going round in circles. Most frustrating is I don't quite understand why one version throws an error, another throws out the wrong number and then whatever you guys come up with will give me the correct answer!

I have a workbook uploaded here which I used for a different question, but majority of the data model is the same : https://onedrive.live.com/?cid=1C0322C61F94C9E9&id=1C0322C61F94C9E9%21125

What I'm trying to do is find staff members that are at risk of burn out (eg consistently working > 8hrs a day)

My main timesheet table (Time) is as such:
StaffID (fk to staff table)
ClientID (fk to client table)
TaskID (fk to task table)
TimesheetDate (fk to date table)
EffortInHrs

my date dim is standard fare, only thing i've added is an EndOfWeekDate (eg EOWDate. sunday is the last day therefore for each date for this week the endofweekdate column = 18-jan. For last week it was 11-Jan etc) and I also have a 'last freshed date' that shows the last time the model was refreshed

A staff member can only record time against a specific task once per day, but they can have multiple tasks across multiple clients for one day

The business rule im working with is
if they've worked more than 8hrs at least once in the last 3 weeks (dates that fall between EOWdate for 'now' minus 3 weeks and EOWdate for 'now' ), then return the sum of EffortInHrs for all clients for every day in the last 3 weeks

I've given up in getting it in one go so i'm trying to cobble together a number of calc columns and measures to get me closer.. but it all seems to stall

eg the following calc column (DailySumOfHrs)

=
CALCULATE (
    SUM ( [EffortInHours] ),
    FILTER (
        ALLEXCEPT ( TIME, 'Time'[TimesheetDate], TIME[StaffID] ),
        DATESBETWEEN (
            'Date'[Date],
            'time'[TimesheetDate],
            'Time'[TimesheetDate]
        )
    )
)

gives me the sum of effort in hrs for the day for each staff member. I couldn't get a measure version of this working. It would either aggregate across all days/staff or give context errors.

So when I tried to take it to the next step to find the max using the same formula

=
CALCULATE (
    SUM ( [EffortInHours] ),
    FILTER (
        ALLEXCEPT ( TIME, 'Time'[TimesheetDate], TIME[StaffID] ),
        DATESBETWEEN (
            'Date'[Date],
            'time'[TimesheetDate],
            'Time'[TimesheetDate]
        )
    )
)

I get errors, i'm guessing because i'm mixing my fact and date tables now and it doesn't know it can use the TimeSheetDate 'context' to filter.

To get the 'last 3 weeks' filter working I added the following to the date dim

=if([EndOfWeek]<[RefreshDateLocal] +7 && [EndOfWeek] > [RefreshDateLocal] -14,1,0)

i then used it in an intermediate measure to get the max at the weekly grain

WeeklyMax :=
IF (
    MAX ( 'Date'[Last3WeeksFilter] ) = 1,
    CALCULATE (
        MAX ( TIME[DailySumOfHrs] ),
        FILTER (
            ALL ( 'Date'[date] ),
            MAX ( 'Date'[EndOfWeek] ) = MAX ( 'Date'[EndOfWeek] )
        ),
        'Date'[Last3WeeksFilter] = 1
    ),
    BLANK ()
)

and finally the measure itself

BurntOutHrs:=if([WeeklyMax]>8,SUM([EffortInHours]),blank())

I am not happy with this at all. There are so many conditions (eg filters added) that need to be satisfied for this to return a sensible result and I need to expose a number of the 'backend' filters that users would need to add to the report to get the correct answer.

As is, it only works for a simple Staff, date, burntouthrs report. If users attempt to create their own pivot table/chart and try to break it down by client or task it doesn't look right at all. I'd also like to be able to count how much stuff are working long hrs but this approach wouldn't support that at all.

I could probably post dozens of my attempted dax that to me logically look like it should work but either aggregate wrong or throw context errors. Unfortunately my current approach to dax still is to try various, almost random variations until one works for whatever reason

I always aim to have a measure that 'just works' in all contexts where it makes logical sense. For me to get that happening here i'd probably need to spend another few hrs trying all kinds of allexcept() conditions for each filter where i want it to work and hope none contradict each other

Thinking about it now, this isn't really a separate measure at all.. i think a better approach would be to add a calc column to the staff table that checks if that staff member has >8hrs assigned for any day (i dont think the default time->staff relationship would allow this though). But i'm still interested on how to accomplish it using the separate 'measure' approach

I'll go spend the rest of today trying to get the flag working in the staff dim!

Thanks!

Jakub @ Adelaide, Australia Blog

Hello,

My PowerPivot report is using quite large data that can be split by subsidiary. I’d like to let user select subsidiary. The selection would trigger change of the PowerPivot’s data connection so that the SQL would change into “SELECT*FROM dbo.ECSPURSL00WHERESubsdiary ='selected subsidiary'”.

The Excel 2013 has Model class that should manage the PowerPivot data sources. The object explorer shows CreateModelWorkbookConnection method, but I can't figure out the VBA syntax. Can anybody help please?

Thank you!

Daniel

Have win 7 Swedish version 64 bit with Office 365 ProPlus 64 bit English version installed on my laptop. A collegue of mine with Swedish Office 32 bit has the same problem.

Since a while back, not sure when, the automatic DAX syntax help is not working as it should. In the Swedish version semicolon is the delimiter between the different part of the formula, and the syntax help has worked perfect with that previously. Now I need to enter the formula with comma as delimiter for the syntax help to work, and then go back and change all the commas to semicolons for the formula to work. Doable but annoying.

Does anyone know why this problem has shown up now?

//Caj

Csten

Hello,

My PowerPivot workbook is deployed in a network folder (no SharePoint) and I am using a small procedure to refresh in the workbook opens  (http://www.vandeputte.org/)

Private Sub Workbook_Open()
    If MsgBox("Do you want to refresh the PowerPivot Model?", vbYesNo) = vbYes Then
        Me.Model.Refresh
    End If
End Sub

Generally, I use my windows authentication and  the macro works just fine. Becuase I am deploying in a secured network folder, I am hoping to have my users (2-3 people), open the workbook and have it refresh. The only hickup is that I can't get Power Pivot or Excel to save my SQL Server Authentication password and checked the "save my password" box, but it won't save it.



I read several post that suggested "checking save" password in the Excel connection, which I did, but I still can't get the application to save the password.  

Any suggestions to refresh on the network is greatly appreciated. 

Thanks

Brent

Hello together,

i have a Problem with a measure field from a Cube in power Pivot.

if i select the measure in the query designer, the measure is correctly shown as a value.

After i Close the query designer and return the power Pivot to Excel sheet,

the measure is shown as a text field so that no sum function is possible.

can u please help me with that issue? 

Hi,

This is a follow up question to an excellent answer from Michael using a calculated column. I’m going to try and use a measure to limit the calculated column, but I using the same theory of his post.

I want to count the rows in the Incident Table by using filters to limit the rows to be counted if they meet the below conditions. I know I need a logical test for each row of the incident table based on the apparatus table’s rows, much like Michael’s solution. But, I want to test for each row in the incident table, counting, but not returning a true or false in the overall measure.

Something like look at each incident row, test for true or false and then count IF the statement is true. Then go to the next incident row and do the same. The aggregation would be the final count of “true” results.

I tried this for MET objective:

=CALCULATE(COUNTROWS(incident),

        apparatus[Incident Response Time] >-1 ||

        apparatus[Incident Response Time] <320,

        uv_901APP_TYPE[Description]="Engine",

        uv_901INCIDENT[Top_Category]="Fire"

         )

And this for Did NOT meet objective

=CALCULATE(COUNTROWS(incident),

        apparatus[Incident Response Time] >320,

        uv_901APP_TYPE[Description]="Engine",

        uv_901INCIDENT[Top_Category]="Fire"

         )

I hope I explained this clear enough, but HERE is a copy of my workbook for further explanation.

Thank you for any help. Brent

Brent

After configuring the synomys in Power BI site, it is no longer possible to open the Power Pivot window. The Excel sheets and Power View tabs still work.

Note: although this is the same error as the other message in this thread, the stack information is clearly different. There are no errors in calculations.

Error: Unable to load the tables in the PowerPivot window.

Details >>

Half Scheidl

Hi,

I experiencing some problems with calculated fields and their decimal separators. When I try multiply a column with data type 'Decimal number' with 0.7 it multplies the amounts with 7 instead of 0.7 (e.g. 18,30*0.7 as ahs result 128,10 if I use 18,30*0,7 a syntax error is raised). When I change the decimal separator to , like 0,7 the error below is raised.

Additional information:

The regional settings are for Number: 123.456.890,12 and MS Excel 2010 is using the system operators.

Product version Microsoft.AnalysisServices.Modeler.FieldList.dll: 10.50.1600.1

============================
Error Message:
============================

The syntax for ',' is incorrect.
The calculated column 'Injection[CalculatedColumn1]' contains a syntax error. Provide a valid formula.

============================
Call Stack:
============================

   at Microsoft.AnalysisServices.Modeler.Storage.DataModelingSandbox.ExecuteCaptureLogAndProcessResults(OperationType type, Boolean cancellable, Boolean raw)
   at Microsoft.AnalysisServices.Modeler.Storage.DataModelingSandbox.ExecuteAMOCode(OperationType type, OperationCancellability cancellable, AMOCode code)
   at Microsoft.AnalysisServices.Modeler.Storage.DataModelingTable.Update(UpdateOptions updateOptions)
   at Microsoft.AnalysisServices.Modeler.Storage.DataModelingView.Update(UpdateOptions updateOptions)
   at Microsoft.AnalysisServices.Modeler.Storage.DataModelingView.ChangeFormulas(IList`1 colIndices, IList`1 names, IList`1 expressions)
   at Microsoft.AnalysisServices.Modeler.ClientWindow.ChangeFormula(TableWidgetPanel currentTable, IList`1 colIndices, IList`1 names, IList`1 expressions, Boolean doFormulaBarCommit)

============================

Greetz,

Eddy

Hi eb

My users are realy looking forward to adding gauges to our power pivot dashboards

currently made of pivot charts.

We are using Excel 2013 updated to october 2014.

As for now i dont see excel offering built in gauges.

Should this change in a near update?

If not would a third party gauge - and any other third party graphical addon for this matter - be supported in a 2013 sharepoint gallery after publishing it??

If not would that be supported if gauge was downloaded from the office app store??

TIA!

Rea

Hi All,

I have deployed one PowerPivot Worksheet containing only data model on SharePoint 2013.

Then I have generated some reports out of it. I mean using that deployed worksheet(as a Data Model) I created some reports (Power-Pivot Link Table as a Data Source) in other Excel files. and then I uploaded these newly created reports on sharepoint.

Now suppose when I open these newly created reports, I want to send parameters with that. so based on these parameters my newly created report should vary.

So how can I send parameters in URL so that my excel will get filtered?

I'm having some difficulty updating a table in my powerpivot data model. The data for the table is stored in a local excel file; if I go toExisting Connections select the connection and do a test then everything is successful. When however I try to refresh that particular table I get the following error:

"We couldn’t refresh the connection. Please go to existing connections and verify they connect to the file or server."

In the past I had this and it was an authentication issue, the files are stored on SharePoint and I wasn't logged in to Sharepoint - I have however checked this and it can't be the issue because I have other tables linked to other excel files in the same location that are refreshing without issue. Can anyone think of what may be causing it?

I have even tried creating a new local excel with the same data, added it to the data model and then tried a refresh and I I get exactly the same issue. 

Are there any known issues with refreshing data from an Excel file if, after the initial import, you add columns to the table in the excel file I was wondering if this might contribute?

Thanks