Posted On: Jun 1, 2022
Amazon SageMaker JumpStart helps you quickly and easily solve your machine learning problems with one-click access to (a) more than 300 popular model collections from TensorFlow Hub, PyTorch Hub, Hugging Face and Gluon CV, and (b) 18 end-to-end solutions that solve common business problems such as demand forecasting, fraud detection and document understanding. The available models can be used for a wide range of machine learning tasks including image classification, object detection, semantic segmentation, instance segmentation, image embedding, text classification, sentence pair classification, question answering, text embedding, text summarization, text generation, machine translation, tabular classification and tabular regression.
Training machine learning models with large data sets can take a long time. Customers often want to improve the quality of a previously trained model when additional training data becomes available. Training the model again with both the old data and new data can take a longer time. Starting today, customers can incrementally train all the JumpStart trained models with new data without training from scratch. This can significantly shorten the training time to reach a better model. This incremental training capability is available through the SageMaker JumpStart UI inside of SageMaker Studio, as well as through python code using SageMaker Python SDK.
Amazon SageMaker JumpStart is available in all regions where Amazon SageMaker Studio is available. To get started with these new models on SageMaker, refer to the JumpStart documentation.