Preparing a Dataset Based on File Data - Amazon QuickSight

Preparing a Dataset Based on File Data

Use the following procedure to prepare a dataset based on text or Microsoft Excel files from either your local network or Amazon S3.

  1. Open a file dataset for data preparation by choosing one of the following options:

  2. (Optional) On the data preparation page, enter a new name into the dataset name box on the application bar.

    This name defaults to the file name for local files. For example, it defaults to Group 1 for Amazon S3 files.

  3. Review the file upload settings and correct them if necessary. For more information about file upload settings, see Choosing File Upload Settings.

    Important

    If you want to change upload settings, make this change before you make any other changes to the dataset. New upload settings cause Amazon QuickSight to reimport the file. This process overwrites all of your other changes.

  4. Prepare the data by doing one or more of the following:

  5. Check the SPICE indicator to see if you have enough capacity to import the dataset. File datasets automatically load into SPICE. The import happens when you choose either Save & visualize or Save.

    If you don't have access to enough SPICE capacity, you can make the dataset smaller by using one of the following options:

    • Apply a filter to limit the number of rows.

    • Select fields to remove from the dataset.

    Note

    The SPICE indicator doesn't update to how much space you save by removing fields or filtering the data. It continues to reflect the SPICE usage from the last import.

  6. Choose Save to save your work, or Cancel to cancel it.

    You might also see Save & visualize. This option appears based on the screen that you started from. If this option isn't there, you can create a new visualization by starting from the dataset screen.

Preparing a Dataset Based on a Microsoft Excel File

Use the following procedure to prepare a Microsoft Excel dataset.

  1. Open a text file dataset for preparation by choosing one of the following options:

    • Create a new Microsoft Excel dataset, and then choose Edit/Preview data. For more information about creating a new Excel dataset, see Creating a Dataset Using a Microsoft Excel File.

    • Open an existing Excel dataset for editing. You can do this from the analysis page or the Your Datasets page. For more information about opening an existing dataset for data preparation, see Editing a Dataset.

  2. (Optional) On the data preparation page, enter a name into the dataset name box in the application bar. If you don't rename the dataset, its name defaults to the Excel file name.

  3. Review the file upload settings and correct them if necessary. For more information about file upload settings, see Choosing File Upload Settings.

    Important

    If it's necessary to change upload settings, make this change before you make any other changes to the dataset. Changing upload settings causes Amazon QuickSight to reimport the file. This process overwrites any changes you have made so far.

  4. (Optional) Change the worksheet selection.

  5. (Optional) Change the range selection. To do this, open Upload Settings from the on–dataset menu beneath the login name at the top right.

  6. Prepare the data by doing one or more of the following:

  7. Check the SPICE indicator to see if you have enough space to import the dataset. Amazon QuickSight must import Excel datasets into SPICE. This import happens when you choose either Save & visualize or Save.

    If you don't have enough SPICE capacity, you can choose to make the dataset smaller using one of the following methods:

    • Apply a filter to limit the number of rows.

    • Select fields to remove from the dataset.

    • Define a smaller range of data to import.

    Note

    The SPICE indicator doesn't update to reflect your changes until after your load them. It shows the SPICE usage from the last import.

  8. Choose Save to save your work, or Cancel to cancel it.

    You might also see Save & visualize. This option appears based on the screen that you started from. If this option isn't there, you can create a new visualization by starting from the dataset screen.