Pixeltable

Pixeltable is a Python library providing a declarative interface for multimodal data (text, images, audio, video). It features built-in versioning, lineage tracking, and incremental updates, enabling users to store, transform, index, and iterate on data for their ML workflows. Data transformations, model inference, and custom logic are embedded as computed columns- Load/Query all data types: Interact with video data at the frame level and documents at the chunk level - Incremental updates for data transformation: Maintain an embedding index colocated with your data - Lazy evaluation and cache management: Eliminates the need for manual frame extraction - Integrates with any Python libraries: Use built-in and custom functions (UDFs) without complex pipelines - Data format agnostic and extensibility: Access tables as Parquet files, PyTorch datasets, or COCO annotations Get started at https://github.com/pixeltable/pixeltable