Automatic Generation of Document Formatting Parameters based on Device Capabilities

Users may view and/or edit documents on a variety of devices, with different screen orientations (portrait/landscape), size, resolution, etc. If a document formatted for viewing on a particular type of device is displayed on a different type of device without taking into account device capabilities, the resulting user experience may require the user to perform cumbersome zoom, pan, or scroll operations. This is because the formatting and visual choices made for the document on the original device may not necessarily be optimal for viewing on the other device. This disclosure describes the use of a trained machine learning model to automatically generate parameters for document formatting to eliminate the need for such manual actions. The parameters include font size, margins, page orientation, etc. The parameters can be provided to the document application which can automatically format the document accordingly. The described techniques thus provide improved document viewing experience without requiring modifications to the document viewing application