Build an AI model

You can either create an experiment with the Experiment wizard or create a blank experiment.

Remember to use a good naming strategy for your versions. Version 1, Version 2, Version 3, etc., becomes a little hard to decipher after a while.

Experiment wizard

The goal of the Experiment wizard is to make it easy to create a ready-to-run experiment. Everything needed will be preset; input and target feature, loss function, activation, runtime settings, i.e., batch size, learning rate, number of epochs.

The idea is to make it easy for DL-non-savvy and platform-non-savvy users to get started quickly on the platform. The wizard takes advantage of the available info, provides a good starting point, and makes sure the user gets a good result with almost no effort.

There are several ways to open the wizard, for example:

  • In the Modeling view, click New experiment.

  • In the Datasets view, click Use in new experiment.

Dataset tab

The first step is to select a dataset. The latest saved dataset and version will be autoselected. Change if you want to.

Based on the input dataset, the wizard will propose a Training subset and a Validation subset. Most often, these will be the subsets that are named something close to Training or Validation, pretty straight forward.

Inputs / target tab

Based on the information in the dataset, the wizard selects features that look like inputs and target. You can off coures change to other inputs or targets.

If you have more than 50 features you need to use the arrows to navigate to the next tab.

Problem type tab

This is where the experiment wizard shines. Based on the input and target feature the wizard will select problem type.

The Problem type together with the input and target features will decide how the wizard-built model will look like. It’s as easy as that.

Example: With English text as input data and Problem type Single-label text classification, the wizard will create a model using a English BERT block.
Example: With images in size 224x224 or above as input data and Problem type Image regression, the wizard will create a model using an EfficientNet block.

Click Create and a ready-to-run experiment will populate the Modeling canvas.

Create button

Blank experiment

If you feel that you know exactly what you’re doing, you can click on Create blank experiment. Then a new experiment will open up with a clean blank Modeling canvas.

Add a block to an experiment

In the Build tab in the Inspector, click on the Block you want to add to the experiment.

The new block will be connected if you select an unconnected block on the canvas and then click to add a new block. Then the new block will appear under the previous block on the canvas. A connection between the two blocks is created automatically.x

The new block will be unconnected if no block has been selected on the canvas. Then drag the new block to where you want it in the model. Connect the new block to another block by using hold-and-drag between the connecting points.

If needed, remove a connector or block by selecting it and then pushing the delete or backward key on the keyboard or clicking the Delete button in the GUI.

Set block parameters

Select the newly added block. Set the parameters for the block. You can find a description of the settings for each block in the Knowledge center.

The Information pop-up will give you information if you need to change a block setting.

Was this page helpful?