> ## Documentation Index
> Fetch the complete documentation index at: https://help.dingtalk.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Create a Dataset from a File

> YiDA datasets support creation from local Excel and CSV files, with no need to create a separate Data Source. Connect files with Forms to quickly extract and process Data. File Fields are processed as strings by default and can be adjusted to types such as Numerical or Date.

Currently supported file types include Excel `.xls` and `.xlsx` files commonly used in organization information systems, as well as CSV (Comma Separated Values) files suited for large-scale text processing. Uploaded file fields are processed as strings by default, and field type adjustments (numerical, date, and more) are supported so you can apply the right data type in each scenario.

## Steps

Follow the steps below to create a dataset from a local file.

1. Sign in to the YiDA workbench and open the App Settings page.

2. Click Data Factory > Dataset.

3. Click Create Dataset > From File.

4. Refer to the following information to select and upload the file.

| **Configuration Item** | **Description**                                                                                                                                                                                                                                                                                                                                                                                                                                |
| ---------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Dataset Name**       | Enter the dataset name, up to 30 characters.                                                                                                                                                                                                                                                                                                                                                                                                   |
| **File Type**          | Select the File Type and sheet you want to upload.                                                                                                                                                                                                                                                                                                                                                                                             |
| **Header**             | Specify whether the uploaded file contains a header.<br />If a spreadsheet file with a header is set as having no header, the header column in the spreadsheet is treated as the first record of data in the dataset.                                                                                                                                                                                                                          |
| **Uploaded File**      | Select the file to upload.<br />The uploaded spreadsheet must meet the following requirements:<br />• **File Type**: Only `.xlsx` **and** `.csv` files are supported.<br />• **File Size**: The file size must not exceed **50M**. For large files, save as a CSV file and upload again.<br />• **Notes**: The Excel content area must **not contain merged cells or formula functions**, and the **header** must **not contain empty** cells. |

5. Configure the dataset field aliases and field types, then click Next when done.

Form fields support four data types: **STRING**, **INTEGER**, **DOUBLE**, and **DATE**.

6. Preview the data to verify it is correct. If the data is correct, click Confirm to complete the creation.