Skip to content

QSPy Outputs and Logs

QSPy automatically generates a variety of outputs and logs to help you track, audit, and reproduce your modeling work. By default, these files are stored in a hidden .qspy folder in your project directory.

What is in the .qspy Folder?

The .qspy folder is created automatically when you run QSPy code. It contains:

  • Model summaries: Markdown files summarizing model structure, parameters, and metadata (output from the Model.markdown_summary function)
  • Model diagrams: Mermaid or image files visualizing model architecture (if enabled by using the ModelMermaidDiagrammer).
  • Run logs: Detailed logs of model construction, macro usage, and simulation runs.
  • Audit trails: Metadata and hashes for reproducibility and version tracking (if enabled by using the ModelMetadataTracker).

This folder is intended to be a central location for all QSPy-generated artifacts, making it easy to review your modeling workflow and share results.

Example Contents

.qspy/
├── model_summary.md
├── model_diagram.mmd
├── logs/
├───|──── qspy.log
├── metadata/
├───|──── model-name__author__short-hash__time.toml
└── ...

Changing the Output Location

By default, QSPy writes all outputs and logs to .qspy in the current working directory. You can change this location using the qspy.config module.

Change the output directory

import qspy.config

# Set a new output directory for QSPy logs and artifacts
qspy.config.set_output_dir("my_outputs/qspy_artifacts")

This will update qspy.config.OUTPUT_DIR, and all new logs and outputs will be written to the specified folder.

Tips

  • Version control: You may want to add .qspy/ to your .gitignore if you do not wish to track logs and outputs in version control.
  • Reproducibility: The logs and metadata in .qspy are useful for reproducing results and tracking model changes over time.
  • Cleanup: You can safely delete the .qspy folder if you want to clear outputs; it will be recreated as needed.

See Also