Concepts
Brief descriptions for each of the main components of our AutoML platform
Datasets
1. Dataset
A dataset is a collection of text, images, or files representing data that will be used for training an AutoML model or as input into a dataset function. Once created, datasets can be edited or updated at any time. For more information on datasets, see Datasets.
2. Snapshot
A snapshot is a point-in-time version of a dataset that is used to train a model. Snapshots are automatically validated on creation to ensure they are suitable for model training. For more information on snapshots, see Snapshots.
Models
1. Model
An AutoML model is a fine-tuned machine learning model, created to perform specific tasks based on the snapshot used to train it. AutoML offers image classification, text classification, and large language models (LLMs). For more information on models and model training, see Models.
2. Deployment
Deployment is the process of preparing your model to support inference requests via the Hive API. AutoML’s simple one-click deployment process automatically creates a Hive Models project that can be used in same way as all of Hive’s pre-trained models. See the Hive Models documentation to learn more about accessing a deployed model.
Updated about 2 months ago