PyLEMS 0.3.01

A Python library for working with the Low Entropy Model Specification language (LEMS)

# PyLEMS

###A [LEMS](http://neuroml.org/lems/)/[NeuroML2](http://neuroml.org/neuroml2.php) simulator written in Python


## Installation
Check the code out of github.

git clone [e-mail]:LEMS/pylems.git

This will create a directory named pylems. Add this directory to your PATH and PYTHONPATH variables.

## Usage

pylems [\<options\>] \<LEMS file\>

### Options
- -I/-include \<path\> - Adds a directory to the model file include search path

## Examples

### *NeuroML* examples (from https://github.com/NeuroML/NeuroML2/tree/development/NeuroML2CoreTypes)

- Example 0 -- Working
- Example 1 -- Working
- Example 2 -- Working
- Example 3 -- Working
- Example 4 -- Not working (Unsupported in PyLEMS: KSChannel)
- Example 5 -- Working
- Example 6 -- Working
- Example 7 -- Working
- Example 8 -- Working
- Example 9 -- Working
- Example 10 -- Working
- Example 11 -- Working
- Example 12 -- Not working (Unsupported in PyLEMS: Property)
- Example 13 -- Working
- Example 14 -- Not working (Unsupported in PyLEMS: Property)
- Example 15 -- Working
- Example 16 -- Working (apart from spikeArray)
- Example 17 -- Working
- Example 18 -- Working


## LEMS elements that do not work
- XPath based parameters - PathParameter
- Assertions

## Tasks TODO
- Implement flattening
- Decouple events from runnables
- Perform dimension-checking on expressions.
- Implement LEMS API over lems.model.* (NeuroML API?)
- Interface with libNeuroML and Pyramidal to export Neuron MOD files
- Export C files (Interface? Steve Marsh’s project?)
- Assertions.
- XPath implementation.
- Implement Runnables from Component types instead of expanded typeless Components (Required for efficient C/C++ code generation, but conflicts with flattening)



## Travis integration

pylems is integrated with the the [Travis Continuous Integration service](http://travis-ci.org/).

[](https://travis-ci.org/LEMS/pylems)