Skip to content
R2X framework logoR2X framework logor2x-reeds

Inspect a saved system

Continue from the fixture tutorial. Use outputs/pacific-system.json and its generated sibling time-series directory. These are Infrasys serialization, not a native PLEXOS, Sienna or upstream model file.

Reload in the project environment

Section titled “Reload in the project environment”

Open python in the tutorial workspace after completing the tutorial:

>>> from r2x_core import System
>>> from r2x_reeds import ReEDSRegion, ReEDSGenerator, ReEDSDemand
>>> system = System.from_json("outputs/pacific-system.json")
>>> assert len(list(system.get_components(ReEDSRegion))) == 11
>>> assert len(list(system.get_components(ReEDSGenerator))) == 673
>>> load = next(item for item in system.get_components(ReEDSDemand) if item.name == "p1_load")
>>> profile = system.get_time_series(load, name="max_active_power")
>>> assert profile.name == "max_active_power"
>>> assert len(profile.data) == 8760
>>> from datetime import datetime, timedelta
>>> from numpy.testing import assert_allclose
>>> assert profile.initial_timestamp == datetime(2012, 1, 1)
>>> assert profile.resolution == timedelta(hours=1)
>>> assert_allclose(profile.data[:3], [7662.1055, 7358.9473, 7142.1055], rtol=1e-6)

For the completed tutorial, expect ReEDSRegion 11; ReEDSGenerator 673, followed by an 8,760-value max_active_power profile. This checks both component reconstruction and attached storage. Compare timestamps and first values with the tutorial. For another dataset, replace the fixture-specific component name rather than silently selecting an unrelated load.

Leave the Python session with exit() before the following shell commands.

Inspect with r2x-cli

Section titled “Inspect with r2x-cli”

Install r2x-cli, then inspect its managed environment:

Terminal window
r2x python show

Install the published package into that CLI environment:

Terminal window
r2x install r2x-reeds

Installation changes the CLI managed environment. It is separate from the project’s uv environment. Serialized objects name their Python module and class, so the matching package must be importable when loading them. Keep compatible Infrasys and r2x-core versions with the environment that wrote the artifact.

For a repeatable CLI check, open python again in the tutorial workspace and create inspect_cli.py. Reload the tutorial artifact in this new session: The script uses sys, the system supplied by r2x read:

>>> from pathlib import Path
>>> from r2x_core import System
>>> from r2x_reeds import ReEDSRegion, ReEDSGenerator, ReEDSDemand
>>> system = System.from_json("outputs/pacific-system.json")
>>> inspection = """from r2x_reeds import ReEDSRegion, ReEDSGenerator, ReEDSDemand
... assert len(list(sys.get_components(ReEDSRegion))) == 11
... assert len(list(sys.get_components(ReEDSGenerator))) == 673
... load = next(item for item in sys.get_components(ReEDSDemand) if item.name == "p1_load")
... profile = sys.get_time_series(load, name="max_active_power")
... assert len(profile.data) == 8760
... print("Verified 11 regions, 673 generators and 8760 demand samples")
... """
>>> _ = Path("inspect_cli.py").write_text(inspection)
>>> exec(inspection, {"sys": system})
Verified 11 regions, 673 generators and 8760 demand samples

Leave Python with exit(), then run this repeatable inspection from the tutorial workspace in your shell:

Terminal window
r2x read outputs/pacific-system.json --exec inspect_cli.py

Expect Verified 11 regions, 673 generators and 8760 demand samples. Omit --exec for an interactive session; type exit() to leave it. The first read may install CLI runtime dependencies. The JSON inspection command was executed with r2x-cli 0.1.15 in an isolated Python 3.12 environment containing the package’s locked dependencies.

For an archive produced by r2x run ... --zip -o outputs/system.zip, pass --zip to r2x read. Do not add --zip when loading the JSON file above. See R2X artifact inspection for the distinction between a serialized system and a native target export.

Recover a failed reload

Section titled “Recover a failed reload”
  • ModuleNotFoundError: install the package owning the recorded class in the loading environment.
  • Missing time-series storage: restore the sibling directory beside the JSON; copying JSON alone is insufficient.
  • Existing output on a rerun: choose a fresh filename or deliberately set overwrite=True in your writer.
  • Counts differ: confirm that you loaded the modified/tutorial artifact and selected the same scenario/years.

Continue with troubleshooting rather than renaming an arbitrary target-format JSON to make it look like Infrasys serialization.