Creating a Report in Power BI Web from a local Power BI Template with Parameters

[Jeff2Jets]

I have a Power BI Template locally on my laptop. In the Power BI template, I have defined parameters for the six .csv files, which allow me to specify the .csv files when I open the template in Power BI Desktop. However, one of the .csv files is > 1 GB, and when I try to use Power BI Desktop to load the .csv files, there is not enough memory on my laptop for the load.

Is there a way to upload the Power BI Template into Power BI Web and then be able to open the template in Power BI Web to be prompted for the location of the six .csv files?

If this is not possible, what are my options?

Please provide specific instructions if possible.

Thanks!

[Shubham_rai955]

There is currently no direct way to upload a Power BI Template (.pbit) to Power BI Web (Power BI Service) and have the service prompt you to select local files for parameterized CSV sources as you would in Power BI Desktop. When you publish a report from Desktop to Power BI Service, you are essentially uploading the data model and queries in a .pbix file, and the Power BI Service cannot access files stored on your local computer for refresh or parameter prompts because it lacks access to your file system.​

What Happens When Uploading a Template or PBIX to Power BI Service

• In Power BI Desktop, templates (.pbit) do prompt for parameters for file locations at initial load.​

• Power BI Service only allows you to change parameter values in published datasets through the dataset settings page. However, file paths must point to locations accessible to the service (e.g., URLs, OneDrive, SharePoint Online).​

• The service itself cannot prompt for or upload files from your local machine, nor change the file source to a local location after publishing.

Workarounds and Solutions

1. Use Cloud Storage

• Upload your large CSV (and the other CSVs if needed) to a cloud source like OneDrive for Business, SharePoint Online, or Azure Blob Storage.

• Edit your parameters in Power BI Desktop so the source paths point to the cloud locations.

• Load a small sample of the large file in Power BI Desktop (using parameters or filters to limit the data for initial load).​

• After publishing your PBIX, go to the dataset’s parameter settings in the Power BI Service and update the parameters if needed, then refresh.​

2. Use Power BI Dataflows

• Create a Dataflow in the Power BI Service and upload your CSV data to the Dataflow (via a cloud location or direct upload).

• Connect your report (PBIX) to the Dataflow as a source, instead of direct file import. This shifts the memory load to the cloud and avoids memory issues on your local machine.​

3. Process Large Files Outside Power BI

• If your CSV is too large, consider pre-processing it: split, filter, or convert it with a script or tool (e.g., to Parquet format) before loading.​

• Load only relevant or manageable slices of data in Power BI Desktop.

[Jeff2Jets]

I think Option #1 is viable, however I would appreciate detailed instructions on how to:

Edit your parameters in Power BI Desktop so the source paths point to the cloud locations.

Is that something you can provide with screenshots, if possible? If not, I understand.

Thanks!

[GeraldGEmerick]

@Jeff2Jets Unless I am very much mistaken, no you can't do what you are saying regarding uploading the template to the service, etc. What I would recommend is to edit your template so that for the 1GB file you filter it to just a few rows. You could even load just the first ten rows for example. Then, you open your template and provide the CSV files. You then upload that to the Service and then edit the query in the service to remove that filter and refresh.

[Jeff2Jets]

Gerald,

Thanks for the reply. Would you please provide detail on how I do the following:

1. What I would recommend is to edit your template so that for the 1GB file you filter it to just a few rows.
2....and then edit the query in the service to remove that filter and refresh.