The scidbpy package requires at least:
We assume an existing installation of SciDB is available. Binary SciDB packages (for Ubuntu 12.04 and RHEL/CentOS 6) and source code are available from http://scidb.org. The examples in this tutorial assume that SciDB is running on a computer with host name “localhost,” at port 8080. If SciDB is not running on localhost, adjust the name accordingly.
The scidbpy package requires installation of a simple HTTP network service called “shim” on the computer that SciDB coordinator is installed on. The network service only needs to be installed on the SciDB computer, not on client computers that connect to SciDB from Python. It’s available in packaged binary form for supported SciDB operating systems, and as source code which can be compiled and deployed on any SciDB installation. See http://github.com/paradigm4/shim for source code and installation instructions.
- tested with version 1.9.
- tested with version 2.7. Required for using the Shim interface to SciDB.
- Pandas (optional)
- tested with version 0.15. Required only for importing/exporting SciDB arrays as Pandas Dataframe objects.
- SciPy (optional)
- tested with versions 0.10-0.12. Required only for importing/exporting SciDB arrays as SciPy sparse matrices.
SciDB-Py Package Installation¶
The latest release of
scidb-py can be installed from the Python package index:
pip install scidb-py==16.9.post1
Install the development package directly from Github with:
pip install git+http://github.com/paradigm4/scidb-py.git@legacy
The legacy release of
scidb-py can be installed from the github “http://github.com/paradigm4/scidb-py”, “legacy” branch by downloading the code and typing:
python setup.py install