Running Resolve#

After saving your input data & case settings (refer below to see how this should look like) (as described in Saving Input Data & Case Settings), you are now ready to run Resolve!

Insert details on input file structure and directories and where reht are saved#

Running Resolve from the Scenario Tool#

As in previous versions of Resolve, users can run cases directly from the Scenario Tool. Below the “Run Resolve Cases” header on the right of that tab, you’ll find a green “Run Resolve Cases Locally” button.

  • On Windows, this will create a new command line window, and you will see Resolve progress.

  • On macOS, we have not yet figured out how to show the command line window as Resolve is running. For now, we recommend macOS users run Resolve from Terminal themselves, as described in the next section.

Running Resolve from Command Line#

If you plan to run Resolve via a command line/terminal, use the following instructions. Running Resolve via the command line gives you more options for how the model is run than are exposed in the Scenario Tool, as discussed below.

  1. In a command line (e.g., Command Prompt), navigate into the ./src/resolve/resolve directory

  2. Activate resolve-env conda environment: conda activate resolve-env

  3. Use the command python run_opt.py to run a case. The run_opt.py script accepts the following arguments:

  • --data-folder: The name of your data folder (if different than the default .\data)

  • --solver-name: The name of the solver to use (e.g., gurobi, cplex, amplxpress, appsi_highs)

  • --raw-results: Save all raw Pyomo model components as CSVs (for detailed model inspection).

  • --symbolic-solver-labels: Enable descriptive variable names in the Pyomo model formulation–helpful for debugging.

Tip: If for the installation process, you had used Pycharm or any other Python software, then the recommended best practice is to run resolve from there after saving the ST as this avoids using macros which might cause computational issues relating to excel based macros as these are deprecated in newer configurations. [Ritvik to rephrase]

Examples:

  • Run all cases listed in ./data/settings/resolve/cases_to_run.csv:

    python run_opt.py

  • To run a single case called Core_25MMT, type the name of the case into the command line:

    python run_opt.py Core_25MMT

  • Run all cases from a different data folder called data-new (listed in ./data-new/settings/resolve/cases_to_run.csv):

    python run_opt.py --data-folder data-new

  • Run all cases using cplex as your solver:

    python run_opt.py --solver-name cplex

Note

Hint: If you’re in your command line and unsure what arguments to pass to run_opt.py, use the command python run_opt.py --help to get help!

Note for E3 Staff

Instructions for running Resolve on ethree.cloud available on the encyclopedia