Table Of Contents

Search

Enter search terms or a module, class or function name.

Installation

Introduction

We recommended that you perform a binary installation. You should not attempt to build SIMA if you are not familiar with compiling software from sources.

On Windows, we recommend using a scientific python distribution for installing the various prerequisites. Recommended distributions are:

For Mac OS X, we recommend installing the prerequisites, especially OpenCV, using a package manager, such as MacPorts.

As an alternative, you can also run SIMA within a Docker container, see docker setup.

Prerequisites

SIMA depends on various freely available, open source software. Whenever possible, we recommend installing these dependencies with your operating system’s or Python distribution’s package manager prior to installing SIMA.

Depending on the features and data formats you wish to use, you may also need to install the following packages:

  • OpenCV >= 2.4.8, required for segmentation, registration of ROIs across multiple datasets, and the ROI Buddy GUI
  • picos >= 1.0.2, required for spike inference (>= 1.1 required for Python 3)
  • pyfftw, allows faster performance of some motion correction methods when installed together with FFTW.
  • h5py >= 2.2.1 (2.3.1 recommended), required for HDF5 file format
  • bottleneck >=0.8, for faster calculations
  • matplotlib >= 1.2.1, for saving extraction summary plots
  • mdp, required for ICA demixing of channels

If you build the package from source, you may also need:

If you want to generate the documentation, you will also need:

If you are using the spike inference feature, we strongly recommend installing MOSEK (free for academic use) which greatly speeds up the inference.

SIMA installation

Linux

The SIMA package can be installed from the python package index:

$ pip install sima --user

Source code can be downloaded from https://pypi.python.org/pypi/sima. If you download the source, you can install the package with setuptools:

$ python setup.py build
$ python setup.py install --user

Windows

The SIMA package can be installed from the python package index:

$ pip install sima

Alternatively, the packaged wheel is available at <https://pypi.python.org/pypi/sima> to be installed with your Python distribution’s package manager.

NOTE: You may need to install shapely separately from the package at: <http://www.lfd.uci.edu/~gohlke/pythonlibs/#shapely>

If building SIMA from source on Windows, you may also need to follow these instructions for compiling the Cython extensions.

Mac OS X

For installing the dependencies, we recommend using MacPorts. If you do not already have XCode installed, downloading XCode from the App Store, and then run the following commands in the Terminal to complete the XCode installation and license agreement:

$ xcode-select --install
$ gcc -v

Next, download and install MacPorts. Then run the following command in terminal to install SIMA and its dependencies:

$ sudo port selfupdate
$ sudo port install python27 py27-numpy py27-scipy py27-scikit-image py27-scikit-learn py27-shapely py27-Pillow py27-matplotlib py27-bottleneck py27-pip py27-h5py opencv +python27
$ sudo port select --set python python27
$ sudo port select --set pip pip27
$ pip install sima --user

Docker

Docker images are pre-built installations that you can run immediately on your local machine. To run SIMA inside a Docker container, first follow the Docker Engine installation instructions for your operating system.

From a docker terminal run the latest SIMA image (it will automatically be downloaded), with:

$ docker run -it --rm --net=host --env="DISPLAY" -v $HOME/.Xauthority:/root/.Xauthority:rw
    -v /PATH/TO/DATA:/data --name sima losonczylab/sima bash

This will give you a shell within the container from which you can run the example workflow script:

$ python /sima/examples/workflow.py

or your own custom scripts on any data you’ve mapped into the container from /PATH/TO/DATA