Start by uploading a data set. It will later be used to train your model. It will be private, and only accessible using the current subscription key.

Symanto Brain works with all types of data sources, including social networks, reviews, news and other. Please make sure you have column names in the first data row and raw text in the first column.

Your data, regardless of if it is for model creation, training, or evaluation purposes, can be uploaded in any of the following formats: .txt, .csv, .xlsx

Add dataset

API Docs: https://api.symanto.net/active-learning/docs#/v2/add_examples_v2_datasets_post

Select DatasetS from the side navigation

The app comes with a list of public default datasets. These are public datasets often used in academia.

2. In order to upload a new dataset, click on the ADD NEW tab

3. Give your dataset a name and hit Enter

4. Drag and drop a file or browse files from your computer to upload

5. Select the text column and click the Upload button

If at any moment, more text instances related to the same dataset domain are collected, we can extend it. This will be possible by means of uploading the file, just as we did before, and giving it the same name. We will be notified about this.

Delete dataset

API Docs: https://api.symanto.net/active-learning/docs#/v2/delete_model_v2_datasets_delete

In case you want to delete a dataset, go to the DELETE tab and select the file you want to delete from the drop-down list.

This action is irreversible and will remove it completely from the app’s DB.

Symanto Brain Knowledge Base

Upload your dataset with text instances

Add dataset

Delete dataset

Next: Create a Zero-shot model