Ocelot is designed for Python 3.4 or higher.
Most Ocelot dependencies are pure Python, and this easy to install, but Ocelot uses lxml. (cytoolz, another compiled library, is optional but highly recommended). Therefore, the easiest way to install Ocelot is using Anaconda. Other installation pathways (e.g. macports, homebrew, Christoph Gohlke’s unofficial binaries should also work as well.
To install using
Anaconda, download the relevant Miniconda installer. Be sure to get Python 3.5, and the 64 bit version.
On OS X, you may have to make the downloaded script executable using something like
chmod +x ~/Downloads/Miniconda3-latest-MacOSX-x86_64.sh.
Run the installation script in the terminal/powershell/command window.
If you have other Python installations on your machine, be sure to answer no to the questions about making this your default Python, and add Miniconda to your path. You will then need to adjust the following commands to the
bin directory of your miniconda installation.
Then, run the following commands to create an Ocelot environment:
conda create -y -n ocelot python=3.6
On Windows, then run:
activate ocelot conda install -y pywin32 conda install -c conda-forge -c cmutel appdirs bw2parameters docopt docutils jinja2 lxml pandas pyprind psutil pytest stats_arrays cytoolz toolz voluptuous wrapt jupyter pip install --no-cache-dir https://github.com/OcelotProject/Ocelot/zipball/master
source activate ocelot conda install -c conda-forge -c cmutel appdirs bw2parameters docopt docutils jinja2 lxml pandas pyprind psutil pytest stats_arrays cytoolz toolz voluptuous wrapt jupyter pip install --no-cache-dir https://github.com/OcelotProject/Ocelot/zipball/master
If you install from pypi, you will have to install the
ocelot-lca package - someone pipped us during development!
If Ocelot is correctly installed, you should be able to run the command line application:
To run Ocelot, you will need an unlinked database in ecospold2 format.