Adding training examples via data import

This function is only available in the Enterprise Solution.

The allowed file formats are .csv, .xls and .xlsx. The maximum file size is 10 MB. the training data has to have a minimum of two columns.

To upload a file you can click on the Bildschirmfoto_2020-10-22_um_10.28.51.png-button . If there is no data inside the classifier yet, it will be located in the middle of the page. If you already have data inside, the button will appear on the upper right corner. You then continue as follows: 

 

1️⃣ Upload a file

In the middle of the page, you can choose and upload a file via drag & drop or by clicking the arrow on the file. You can also download and edit and example file under Bildschirmfoto_2020-10-22_um_10.31.08.png.

Bildschirmfoto_2020-10-22_um_10.30.32.png

 

2️⃣ Choose a category and text

Next, you have to choose which column contains the text and which column contains the category: 

Bildschirmfoto_2020-10-22_um_10.33.56.png

 

3️⃣ Remove header from columns

Often, the first row in a table contains a header, which is also the case in the example above. By selecting the following option the headline won’t be imported as data:

Bildschirmfoto_2020-10-22_um_10.35.40.png

After selecting this option, the top row should appear crossed out: 

Bildschirmfoto_2020-10-22_um_10.36.46.png

If the top row isn’t a header but a training example, this step can be skipped.

 

4️⃣ Confirm upload

By clicking on the Bildschirmfoto_2020-10-22_um_10.38.08.png-button, the training data will be uploaded and the import will be completed. 

Was this article helpful?
0 out of 0 found this helpful