Quickstart

This page gets you from clone to first successful Res-IRF4 run.

1. Prerequisites

2. Setup

git clone https://github.com/CIRED/Res-IRF4.git
cd Res-IRF4
conda env create -f environment.yml
conda activate Res-IRF4

The expected conda environment name is Res-IRF4 (from environment.yml).

3. Run your first simulation

Run from the repository root:

python -m project.main -c project/config/config.json

4. Check that the run succeeded

A successful run creates a timestamped folder under project/output/:

ls -1 project/output | tail -n 5

Inside the latest run folder, you should see:

  • summary_run.pdf

  • one directory per scenario

  • scenario-level output.csv files

  • img/ folders with generated figures

Useful CLI flags

Flag

Description

Default

-c, --config

Path to configuration file

project/config/test/test.json

-y, --year

Override simulation end year

unset

-cpu, --cpu

Number of CPUs for parallel runs

6

For batch execution of multiple configs in a folder:

python -m project.runs -d project/config/policies/realistic

Edit a scenario quickly

Main config files are in project/config/.

High-impact fields:

  • "end": simulation horizon (for example 2050)

  • "file": inherit from a baseline config

  • "policies": inject policy sets

  • "simple": run simplifications ("quintiles": true, limited heating systems) for faster iterations

Performance notes

  • Memory can reach about 2 GB per scenario depending on detail level.

  • On typical laptops, keep parallel runs to about 2-3 scenarios at once.

See also