Skip to main content

AutoCalibration tool written in Python

Project description

# parsac

parsac (formerly acpy) is a Python-based tool for sensitivity analysis and auto-calibration in parallel. It is designed for analysis of models that take significant time to run. For that reason, it focuses on storing and exploiting every single model result, and performing model runs in parallel on either a single machine or on computer clusters. It works with models that are run by calling one binary, that use text-based configuration files based on YAML or Fortran namelists, and that write their output to NetCDF.


## Installation

pip install parsac –user

Remove –user to install in the system’s shared Python directory (not recommended). Some systems have multiple versions of pip, e.g., pip for Python 2, pip3 for Python 3. Make sure you use the command that corresponds to the Python version you want to install into.

### Dependencies

parsac supports parallel simulations through [Parallel Python]( This package supports Python 2 out of the box (pip install pp –user), but its Python 3 version is currently in beta. To install pp in Python 3, [download the zip file with the Python 3 port of Parallel Python](, extract its contents, go to the contained directory and open a command prompt there, then run python install.

parsac uses [SALib]( for sensitivity analysis. Typically, this can be installed with pip install SALib –user. If you are using [the Anaconda Python distribution](, you can instead do conda install SALib (you may need to add -c conda-forge).

## Known issues

  • On Windows, parallel runs may finish with several “ERROR: The process “xxx” not found.” messages. These are harmless and can be ignored - the analysis has completed successfully and all results have been correctly processed.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distribution

parsac-0.5.7-py2.py3-none-any.whl (332.2 kB view hashes)

Uploaded py2 py3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page