Skip to content

python module for performing blocking analysis on data containing serial correlations

License

Notifications You must be signed in to change notification settings

jsspencer/pyblock

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

pyblock

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 3.6-3.9 and should work on any other version supported by pandas.

CI build status Documentation Status

Documentation

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

Installation

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 flit
$ flit install --symlink

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.

License

Modified BSD license; see LICENSE for more details.

Please cite pyblock, James Spencer, http://github.com/jsspencer/pyblock if used to analyse data for an academic publication.

Author

James Spencer, Imperial College London

Contributing

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!

Acknowledgments

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.

About

python module for performing blocking analysis on data containing serial correlations

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •  

Languages