Download notebook

Start with the packaged IOT; use your parsed db in the same calls.

import mario

db = mario.load_test("IOT")
sectors = db.get_index("Sector")
production = db.query("X", scenarios="baseline")

Result · sectors is a list of native labels; production is a pandas DataFrame indexed by region and sector.

Sets, labels and scenarios

Labels and sets

db.sets lists the sets. db.get_index("all") returns their members. Use db.search("Agr") to find labels without knowing their exact spelling.

Scenarios and matrices

db.scenarios lists the scenarios. db.list_matrices(scenario="baseline") shows what is stored. db.available_matrices() lists quantities the runtime can resolve.

Read data

Z = db.Z                          # Baseline transactions.
e = db.query("e")                 # Calculate if missing.
selected = Z.loc[("Reg1", "Sector", "Agriculture")]

Dotted access and query() can trigger a calculation. They are more than lookups of already stored matrices. See Calculate matrices for control over requested targets and scenarios.

SUTs: activities and commodities

sut = mario.load_test("SUT")
activities = sut.get_index("Activity")
commodities = sut.get_index("Commodity")
use = sut.query("U")
commodity_output = sut.query("Xc")

The inspection pattern stays the same; SUTs have separate activity and commodity sets and split matrix blocks. See Supply and Use Tables.

BONSAI and other regional classifications

When membership varies by region, get_index() returns {region: [members]}. Ask for one region or the union explicitly:

local = db.get_index("Sector", region="IT")
union = db.get_index("Sector", flat=True)

These calls require a database containing IT. A label in the union need not exist in every region. search() accepts the same selectors.

Continue with named groups, classification concordances or exporting your model.