Setting up the tutorial notebooks

The tutorial notebooks run against a local database. Each writes only to databases named tutorial_* and can be run again.

Notebook	For
Getting started	Anyone new: register a lab's mice, sessions and files on the shared tables, add the lab's own tables, query across them.
An existing lab: MLAI and MouseAR	Labs with an existing pipeline: the Mathis lab's real code for its lab layer, automatic ingestion and legacy tables.

New to DataJoint?

These notebooks assume you know DataJoint's basics: schemas, tables, inserts and queries. SCENE pipelines uses DataJoint 2.2 or later; its official tutorials teach the basics:

DataJoint tutorial	Covers
First Pipeline	Tables and dependencies, inserts, and the four core query operations
Schema Design	Table tiers, primary keys and dependencies
Data Entry	Inserting, updating and deleting
Queries	Restriction, projection, joins and aggregation
Computation	Computed tables and populate()

The DataJoint tutorials also include domain examples, such as calcium imaging and electrophysiology. Its how-to guides cover single tasks, and its concepts cover the data model.

Running them locally

From the repository root:

pip install -e .
make init        # once: creates .env
make db_up       # starts MySQL

The notebooks are in examples/tutorials/. Open them from that folder, so that tutorial_utils can be imported. tutorial_utils.drop_tutorial_schemas() removes the tutorial databases to start over.

Running them on Google Colab

Each notebook has an Open in Colab badge at the top. On Colab, run the first code cell: it installs MySQL and the package in the Colab machine. Nothing is needed on your own computer, and the database is gone when the Colab session ends.