Skip to main content

Reblocking analysis tools for correlated data

Project description

pyblock is a python module for performing a reblocking analysis on serially-correlated data.

The algorithms implemented in pyblock are not new; please see the documentation for references.

pyblock is compatible with (and tested on!) python 2.7 and python 3.3-3.4 and should work on any other version supported by pandas.


Documentation and a simple tutorial can be found in the docs subdirectory and on readthedocs.


pyblock can be used simply by adding to $PYTHONPATH. Alternatively, it can be installed using distutils by running:

$ pip install /path/to/pyblock

where /path/to/ is the (relative or absolute) path to the directory containing pyblock. To install an editable version (useful for development work) do:

$ pip install -e /path/to/pyblock

pyblock can also be installed from PyPI:

$ pip install pyblock

pyblock requires numpy and (optionally) pandas and matplotlib. Please see the documentation for more details.


Modified BSD license; see LICENSE for more details.

Please cite pyblock, James Spencer, if used to analyse data for an academic publication.


James Spencer, Imperial College London


Contributions are extremely welcome, either by raising an issue or contributing code. For code contributions, please try to follow the following points:

  1. Divide commits into logical units (e.g. don’t mix feature development with refactoring).

  2. Ensure all existing tests pass.

  3. Create tests for new functionality. I aim for complete test coverage. (Currently the only function not tested is one that creates plots.)

  4. Write nice git commit messages (see Tim Pope’s advice.)

  5. Send a pull request!


Will Vigor (Imperial College London) pointed out and wrote an early implementation of the algorithm to detect the optimal reblock length.

Tom Poole (Imperial College London) contributed code to handle weighted averages.

The HANDE FCIQMC/CCMC development team made several helpful comments and suggestions.

Project details

Download files

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

Source Distribution

pyblock-0.6.tar.gz (15.9 kB view hashes)

Uploaded Source

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