"""
Module containing SimConfig class including simulation configuration and methods
"""
# required to make json saving work in Python 2/3
try:
to_unicode = unicode
except NameError:
to_unicode = str
from collections import OrderedDict
from .dicts import Dict, ODict
# ----------------------------------------------------------------------------
# SIMULATION CONFIGURATION CLASS
# ----------------------------------------------------------------------------
[docs]
class SimConfig(object):
"""
Class to hold simulation configuration parameters
"""
def __init__(self, simConfigDict=None):
# Simulation parameters
self.duration = self.tstop = 1 * 1e3 # Duration of the simulation, in ms
self.dt = 0.025 # Internal integration timestep to use
self.hParams = Dict({'celsius': 6.3, 'v_init': -65.0, 'clamp_resist': 0.001}) # parameters of h module
self.coreneuron = False # use CoreNEURON to run the simulation (alpha version)
self.dump_coreneuron_model = False # dump CoreNEURON model
self.random123 = False # enable Random123 random number generator
self.cache_efficient = False # use CVode cache_efficient option to optimize load when running on many cores
self.gpu = False # enable GPU execution in CoreNEURON
self.cvode_active = False # Use CVode variable time step
self.use_fast_imem = False # use CVode fast_imem to record membrane voltage via i_membrane_
self.cvode_atol = 0.001 # absolute error tolerance
self.seeds = Dict(
{'conn': 1, 'stim': 1, 'loc': 1, 'cell': 1}
) # Seeds for randomizers (connectivity, input stimulation, cell locations, cell and synMech params)
self.rand123GlobalIndex = (
None # Sets the global index used by all instances of the Random123 instances of Random
)
self.createNEURONObj = True # create runnable network in NEURON when instantiating netpyne network metadata
self.createPyStruct = True # create Python structure (simulator-independent) when instantiating network
self.addSynMechs = True # whether to add synaptich mechanisms or not
self.includeParamsLabel = True # include label of param rule that created that cell, conn or stim
self.gatherOnlySimData = False # omits gathering of net+cell data thus reducing gatherData time
self.compactConnFormat = (
False # replace dict format with compact list format for conns (need to provide list of keys to include)
)
self.connRandomSecFromList = True # select random section (and location) from list even when synsPerConn=1
self.distributeSynsUniformly = True # locate synapses at uniformly across section list; if false, place one syn per section in section list
self.pt3dRelativeToCellLocation = True # Make cell 3d points relative to the cell x,y,z location
self.invertedYCoord = (
True # Make y-axis coordinate negative so they represent depth when visualized (0 at the top)
)
self.allowSelfConns = False # allow connections from a cell to itself
self.allowConnsWithWeight0 = True # allow connections with weight 0
self.oneSynPerNetcon = (
True # create one individual synapse object for each Netcon (if False, same synpase can be shared)
)
self.saveCellSecs = True # save all the sections info for each cell (False reduces time+space; available in netParams; prevents re-simulation)
self.saveCellConns = (
True # save all the conns info for each cell (False reduces time+space; prevents re-simulation)
)
self.timing = True # show timing of each process
self.saveTiming = False # save timing data to pickle file
self.printRunTime = False # print run time at interval (in sec) specified here (eg. 0.1)
self.printPopAvgRates = False # print population avg firing rates after run
self.printSynsAfterRule = False # print total of connections after each conn rule is applied
self.verbose = False # show detailed messages
self.progressBar = 2 # (0: no progress bar; 1: progress bar w/ leave = False; 2: progress bar w/ leave = True)
# Recording
self.recordCells = [] # what cells to record traces from (eg. 'all', 5, or 'PYR')
self.recordTraces = {} # Dict of traces to record
self.recordCellsSpikes = -1 # cells to record spike times from (-1 to record from all)
self.recordStim = False # record spikes of cell stims
self.recordLFP = [] # list of 3D locations to record LFP from
self.recordDipole = False # record dipoles using lfpykit method
self.recordDipolesHNN = False # record dipoles using HNN method
self.saveLFPCells = False # Store LFP generated individually by each cell
self.saveLFPPops = False # Store LFP generated individually by each population
self.saveDipoleCells = False # Store LFP generated individually by each cell
self.saveDipolePops = False # Store LFP generated individually by each population
self.recordStep = 0.1 # Step size in ms to save data (eg. V traces, LFP, etc)
self.recordTime = True # record time step of recording
# Saving
self.simLabel = '' # name of simulation (used as filename if none provided)
self.saveFolder = '' # path where to save output data
self.filename = (
'model_output' # Name of file to save model output (if omitted then saveFolder+simLabel is used)
)
self.saveDataInclude = ['netParams', 'netCells', 'netPops', 'simConfig', 'simData']
self.timestampFilename = False # Add timestamp to filename to avoid overwriting
self.savePickle = False # save to pickle file
self.saveJson = False # save to json file
self.saveMat = False # save to mat file
self.saveCSV = False # save to txt file
self.saveDpk = False # save to .dpk pickled file
self.saveHDF5 = False # save to HDF5 file
self.saveDat = False # save traces to .dat file(s)
self.backupCfgFile = [] # copy cfg file, list with [sourceFile,destFolder] (eg. ['cfg.py', 'backupcfg/'])
# error checking
self.validateNetParams = (
False # whether to validate the input parameters (will be turned off if num processors > 1)
)
# self.checkErrors = False # whether to validate the input parameters (will be turned off if num processors > 1)
# self.checkErrorsVerbose = False # whether to print detailed errors during input parameter validation
# self.exitOnError = False # whether to hard exit on error
# Analysis and plotting
self.analysis = ODict()
# fill in params from dict passed as argument
if simConfigDict:
for k, v in simConfigDict.items():
if isinstance(v, OrderedDict):
setattr(self, k, ODict(v))
elif isinstance(v, dict):
setattr(self, k, Dict(v))
else:
setattr(self, k, v)
def __getitem__(self, k):
try:
return object.__getattribute__(self, k)
except:
raise KeyError(k)
def __setitem__(self, k, v):
try:
setattr(self, k, v)
except:
raise KeyError(v)
[docs]
def save(self, filename):
import os
basename = os.path.basename(filename)
folder = filename.split(basename)[0]
ext = basename.split('.')[1]
# make directories if they do not already exist:
try:
os.makedirs(folder, exist_ok=True)
except Exception as e:
print('%s: Exception: %s,' % (os.path.abspath(__file__), e))
raise SystemExit('Could not create %s' % (folder))
dataSave = {'simConfig': self.__dict__}
# Save to json file
if ext == 'json':
from .. import sim
print(('Saving simConfig to %s ... ' % (filename)))
sim.saveJSON(filename, dataSave)
[docs]
def setParam(self, param, value):
self.__dict__[param] = value
[docs]
def addAnalysis(self, func, params):
self.analysis[func] = params
[docs]
def todict(self):
from ..sim import replaceDictODict
return replaceDictODict(self.__dict__)
[docs]
def validateDataSaveOptions(self, printWarning=True):
if any(
[self.savePickle, self.saveJson, self.saveMat, self.saveCSV, self.saveDpk, self.saveHDF5, self.saveDat]
):
return True
if printWarning:
print(
"Warning: data won't be saved. No output format specified (consider sim.cfg.savePickle, sim.cfg.saveJson etc.)"
)
return False