Skip to content
forked from iiasa/ixmp

The ix modeling platform for integrated and cross-cutting scenario analysis

License

Notifications You must be signed in to change notification settings

scparkinson/ixmp

 
 

Repository files navigation

The ix modeling platform (ixmp)

Overview

The ix modeling platform (ixmp) is a data warehouse for high-powered scenario analysis, with interfaces to Python and R for efficient scientific workflows and effective data pre- and post-processing, and a structured database backend for version-controlled data management.

This repository contains the core and application programming interfaces (API) for the ix modeling platform (ixmp), as well as a number of tutorials and examples for a generic model instance based on Dantzig's transport problem.

License

Copyright 2017 IIASA Energy Program

The platform package is licensed under the Apache License, Version 2.0 (the "License"); you may not use the files in this repository except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0.

Please refer to the NOTICE for details and the user guidelines.

Documentation and tutorial

A documentation the ix modeling platform and the MESSAGEix framework is automatically created from the documentation of the Python and R API packages. The online documentation is synchronyzed with the contents of the master branch of the repositories www.github.com/iiasa/ixmp and www.github.com/iiasa/message_ix.

There are a number of tutorials to get started with ixmp. You may want to try the tutorial/transport...

Follow the instructions in doc/README for building the ixmp documentation including the scientific programming API manuals on your local machine.

Dependency Installation

General

  1. Install the Java Development Kit (Java SE 8) and set the environment variable JAVA_HOME per the JDK website instructions; if JAVA_HOME does not exist, add as new system variable. At this point, ixmp is not compatible with JAVA SE 9.

  2. Update your PATH environment variable to point to the JRE binaries and server installation (e.g., C:\Program Files\Java\jdk[YOUR JDK VERSION]\jre\bin\, C:\Program Files\Java\jdk[YOUR JDK VERSION]\jre\bin\server).

    Do not overwrite the existing `PATH` environment variable, but add to the list of existing paths.
  3. Install the latest version of GAMS, otherwise, the export to GDX may not work properly (visit www.gams.com). If you only have a license for an older verson of GAMS, install both versions. Note that for using the integrated MESSAGEix-MACRO model, it is important to install a GAMS version >= 24.8.1.

  4. Update your PATH environment variable to point to the GAMS installation (e.g. C:\GAMS\win64\24.8); again do not overwrite existing PATH but rather add to end.

  5. Install some version of Python (2.7 is supported, but 3.6 or higher is recommended). Anaconda is a good choice for users not yet familiar with the language (during installation select add anaconda to PATH system variable)

  6. Install a Windows C++ Compiler

  7. Install a version of git, (see, e.g., the website)

  8. In a command prompt, execute the following two lines to install all Python dependencies for running the unit tests and building the auto-documentation for the Python interface and the MESSAGEix GAMS code:

    pip install cython numpy pandas "pytest>=3.0.6" "JPype1>=0.6.2" 
    pip install sphinx sphinxcontrib.bibtex sphinxcontrib-fulltoc numpydoc cloud_sptheme
    

Additional dependencies for R users

  1. Install packages rJava , devtools and optparse via the R package manager

  2. For working with Jupyter notebooks using R, install the IRkernel

Installing the ix modeling platform

  1. Fork this repository and clone the forked repository (<user>/ixmp) to your machine. To fork the repository, look for the fork button in the top right at iiasa/ixmp. Add iiasa/ixmp as upstream to your clone.

    We recommend GitKraken for users who prefer a graphical user interface application to work with Github (as opposed to the command line).

Windows Users

  1. Double click on install.bat in the local folder in which you saved your branch. This will also install the R interface.

*nix Users

  1. To install the Python interface and run the unit tests, execute the following command in a command prompt:

    python setup.py install
    py.test tests
    
  2. To install the R interface, as an additional step , in a command prompt: locate in the ixmp folder where the file install.bat is; type:

    rscript rixmp/build_rixmp.R [--verbose]
    

Connecting to an ixmp database instance

  1. To connect to a central database instance, you need a database properties file and place it in the ixmp/config folder. Refer to the README.md in that folder for details.

    Connecting to a local (file-based) HSQL database instance for testing and developing small model instances does not require a properties file!

Notes and Warnings

  1. For Anaconda users, jpype may not install correctly using the pip command. So, if install.bat cannot find jpype, use conda to install it by running the following command in a command prompt:

    conda install -c conda-forge jpype1
    

    To check if it works now, run the following command and you should get the jpype version number 0.6.2 (or higher).

    python -c "import jpype; print(jpype.__version__)"
    
  2. For Anaconda users experiencing problems during installation of ixmp, Anaconda might not have been added to the PATH system variable properly. So, if install.bat just opens and collapses again, check if:

    C:\[YOUR ANACONDA LOCATION]\Anaconda3;
    C:\[YOUR ANACONDA LOCATION]\Anaconda3\Scripts;
    C:\[YOUR ANACONDA LOCATION]\Anaconda3\Library\bin;
    

    are all part of the PATH system variable. If they are not there, add them.

About

The ix modeling platform for integrated and cross-cutting scenario analysis

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 36.5%
  • Jupyter Notebook 23.3%
  • PLSQL 19.6%
  • R 16.7%
  • GAMS 3.2%
  • Batchfile 0.4%
  • Shell 0.3%