Training with Generate Code

What is Generate Code?

Code generation is a function that automatically generates deep learning model training code. When you upload training data and set the required hyperparameter values for deep learning algorithms, DS2.ai automatically returns code for AI model training, and as soon as you run that code in the Jupyter environment, AI model training begins.

Models Trainable with Generate Code

Structured Data Category Classification

You can predict grades such as A/B/F, Yes or No, or set results or discontinuous values such as specific brands.

pageClassification

Structured Data Regression

It is a function that predicts continuous results in the form of numbers, and can be used in a way that represents the risk in a particular field as a % number, scores in a particular field, or predicts scores.

pageRegression

Object Detection

You can learn images using data that contain labeling in the form of images and JSON, classify which objects are in the image, and distinguish the location of the objects.

pageObject Detection

Using the Generate Code

1. Setting Up the Model Training Options

1. Set the Training Method to Generate Code.

2. Set other options for model training. (For more information, see Training > Training the Model > Model Training Options.)

2. Setting Up the Hyperparameters

You can adjust the hyperparameter values directly. You can enter multiple values by pressing Enter in each input window.

Do you have any other questions? [email protected]