BigML treats images as any other input field. This unique implementation allows you to use image data alongside text, categorical, numeric, date-time, and items data types as input to create any Machine Learning model available in our platform, both supervised and unsupervised, providing ease of deployment, immutability, traceability, and programmability also for images.
You will be able to work with your image data from the BigML Dashboard, via the API, or with WhizzML for end-to-end automation. To make this possible, we are bringing composite sources to the platform. A composite source is a collection of component sources supporting multiple formats, including not only tables, but also individual images. You can then create a dataset from any collection of images, and use them for modeling purposes, as any other dataset. This streamlines your image dataset management and allows combining image inputs with other fields, such as labels for classification, numeric features extracted from the image, or other metadata.
|