Creating your custom model is the bread and butter of what we do. Let's take a step-by-step look at all the different options you have.

You have two options on how to get started - either go with a template (a pre-trained model) or create your own AI block, trained on your data. You can naturally weigh the pros and cons - but we'll guide you through the latter option here.

Creating the block

You can start by using the button on the right-hand corner, or use a shortcut on the "+" next to the AI block overview.

Create an AI block on Levity

This will lead you to an empty template where you'll:

  • Name your AI block for a better overview. This is the name you'll see in the overview on the left-hand menu.

  • Create your labels. This is the step where you should think about what types of classes you want to have - what information you are trying to gain with classification. Do so by clicking on "+ new label", name it, and hit enter.

Starting your classification process on Levity

If you've made a mistake in naming your labels, or you need to delete any of them:

  • Hover over the label you need to change

  • Click on the 3 dots that appear left of the label

  • Select the action

Change your labels on Levity

You're now ready for the next steps!

Adding Learning Data

The next step is feeding the model with some data to learn from. Think of this process as training a new employee - in this case regarding Email Classification, you'd need to give the model examples of what spam or urgent emails look like to you.

Click on Add data

Add data to a model

You now have a couple of options - emails naturally mean that you either upload a CSV or a TXT file with examples of said classes (but in case you're training an image or document classifier, other file types are supported as well).

adding labeled and unlabeled data to your model

Depending on your processes, you can choose to either upload:

  • Add data to label: A folder or bulk of data to separate labels one by one: Select the label on the menu and upload your files.

  • Add organized data: Add a CSV file where one column represents your data and the other column represents your labels.

You can find a more detailed section on data here.

Training

Once you have created more than 2 labels and added more than 20 data points to each label, you can train your model.

You'll see a large button to start training. This process could take anywhere up to 10 minutes, depending on the amount and type of data you have, and your internet connection. In case you're running into any problems, you can do a quick troubleshooting yourself.

Once the training is finished, you'll get a performance score. This is the first indicator of how well your model performs in "real life". You'll need to evaluate the score before continuing.

Testing & Human review

Before you'll set your model up in an automated workflow, you can test it out either by uploading a file from your computer or copying a URL of a file.

You'll get a prediction and the confidence rate - this will give you a good indication of how your model behaves in different situations.

And by the way - Congrats! You now have a working AI block!

You now have a couple of different options here to make sure your model does exactly what you want it to do by:

Did this answer your question?