Datasaur Dinamic

Datasaur Dinamic enables data labeling to train a model natively. It creates a ready-to-use model directly from Datasaur.

Introduction

Datasaur Dinamic is an ideal solution for users who prefer to obtain a trained model directly rather than labeled data. It is integrated with Amazon SageMaker and Hugging Face AutoTrain. It supports row-based tasks with Amazon SageMaker and Hugging Face AutoTrain, and token-based tasks exclusively with Hugging Face AutoTrain.

Key Features

Using Datasaur Dinamic offers several benefits:

1. Automated model training: Streamline the process of training machine learning models using labeled data.

2. Efficient labeling: Use the trained models for ML-assisted labeling, improving the efficiency and accuracy of future labeling tasks.

3. Integration with applications: Deploy the trained models directly into applications, enhancing functionality with precise data insights.

Supported Service Providers

Please ensure that you have a minimum of 10 rows or sentences of labeled data for effective utilization of Datasaur Dinamic.

Service Providers	Supported Project Type	Configurations
Amazon SageMaker	Row Based	Amazon SageMaker Configuration
Hugging Face Auto Train	Row Based and Span Based	Hugging Face Configuration