• N-dimensional labelled array objects to store and manipulate multi-dimensional data
• I/O functions for reading and writing arrays in different formats: CSV, Microsoft Excel, HDF5, pickle
• Arrays can be grouped into Session objects and loaded/dumped at once
• User interface with an IPython console for rapid exploration of data
• Compatible with the pandas library: LArray objects can be converted into pandas DataFrame and vice versa.

The easiest route to installing larray is through Conda. For all platforms installing larray can be done with:

conda install -c gdementen larray

This will install a lightweight version of larray depending only on Numpy and Pandas libraries only. Additional libraries are required to use the included graphical user interface, make plots or use special I/O functions for easy dump/load from Excel or HDF files. Optional dependencies are described below.

Installing larray with all optional dependencies can be done with :

conda install -c gdementen larrayenv

You can also first add the channel gdementen to your channel list :

conda config --add channels gdementen

and then install larray (or larrayenv) as :

conda install larray

The latest release of LArray is available from https://github.com/liam2/larray.git

Once you have satisfied the requirements detailed below, simply run:

python setup.py install

• Python 2.7, 3.4, 3.5, or 3.6
• numpy (1.10.0 or later)
• pandas (0.13.1 or later)

• pytables: for working with files in HDF5 format.
• xlrd: for reading data and formatting information from older Excel files (ie: .xls)
• openpyxl: recommended package for reading and writing Excel 2010 files (ie: .xlsx)
• xlsxwriter: alternative package for writing data, formatting information and, in particular, charts in the Excel 2010 format (ie: .xlsx)
• larray_eurostat: provides functions to easily download EUROSTAT files as larray objects. Currently limited to TSV files.

LArray includes a graphical user interface to view, edit and compare arrays.

• pyqt (4 or 5): required by larray-editor (see below).
• pyside: alternative to PyQt.
• qtpy: required by larray-editor. Provides support for PyQt5, PyQt4 and PySide using the PyQt5 layout.
• larray-editor: required to use the graphical user interface associated with larray. It assumes that qtpy and pyqt or pyside are installed. On windows, creates also a menu LArray in the Windows Start Menu.

• matplotlib: required for plotting.

The official documentation is hosted on ReadTheDocs at http://larray.readthedocs.io/en/stable/

• Report bugs, suggest features or view the source code on GitHub.
• For questions, ideas or general discussion, use the mailing list.