Reference

LaminDB: Manage R&D data & analyses.

Import the package:

import lamindb as ln

The central class of the API is File, a wrapper for files, on-disk (zarr, etc.) and in-memory data objects (DataFrame, AnnData, etc.).

File([data, key, name, run, format, ...])	Files: serialized data objects.
Folder([path, id, name, key, storage_id, files])	Folders: collections of files.

Track runs of data transformations:

Run(*[, id, name, load_latest, external_id, ...])	Runs of data transforms.
Transform(*[, id, version, type, title, ...])	Data transformations.

Track data by feature sets:

Features([iterable, field, id, type, files, ...])

Feature sets.

Query & manipulate data:

select(*entity, **fields)	Query data.
add(record, **fields)	Insert or update data records.
parse(iterable, field[, from_bionty])	Parse a dataset column based on a SQLModel entity field.
delete(record[, delete_data_from_storage])	Delete metadata records & files.

Manipulate data with open session:

Session()

Database session.

Utility functions:

track([transform, new_run, notebook_path, ...])	Track Transform & Run records for a notebook or pipeline.
view([n, schema])	View data.

Basic entities:

User(*, id, email, handle[, name, updated_at])	User accounts.
Project(*[, id, created_by_id, updated_at])	Projects.
Storage(*[, id, type, region, updated_at, ...])	Storage locations, often object storage buckets.

Schema - entities and their relations:

schema

Schema tools & overview.

Setup:

setup

LaminDB setup.

Developer API:

link	Link tables.
context()	Global run context.
settings()	Settings.
types	Types.
dev	Developer API.