Page MenuHomec4science

parametersSetup.py
No OneTemporary

File Metadata

Created
Wed, Jun 29, 14:32

parametersSetup.py

''' file with all parameters'''
import numpy as np
import pickle
Dataset='01_CHBMIT' #'01_CHBMIT', '01_iEEG_Bern'
patients =['01','02','03','04','05','06','07','08','09','10','11','12','13','14','15','16']
class GeneralParams:
#LABEL SMOOTHING PARAMETERS
seizureStableLenToTest=5 #in seconds - window for performin label voting
seizureStablePercToTest=0.5 # 50% of 1 in last seizureStableLenToTest values that needs to be 1 to finally keep label 1
distanceBetween2Seizures=30 #in seconds - if seizures are closer then this then they are merged
timeBeforeSeizureConsideredAsSeizure=30 #in seconds - if seizure starts bit before true seizure to still consider ok
numFPperDayThr=1 #for additional meausre of performance what number of FP seizures per days we consider ok
toleranceFP_befSeiz=10 #in sec
toleranceFP_aftSeiz=30 #in sec
patients=patients #on which subjects to train and test
plottingON=0 #determines whether some additional plots are plotted
PersGenApproach='personalized' #'personalized', 'generalized' approaches
#for generalized model defining iterations for CV (which subj are in test)
#CViterations_testSubj=[[0,1,2,3],[4,5,6,7],[8,9,10,11],[12,13,14,15],[16,17,18,19],[20,21,22,23]]
class SigInfoParams:
chToKeep=np.array([0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17]) #which channels to keep
channels = ['FP1-F7', 'F7-T7', 'T7-P7', 'P7-O1',
'FP1-F3', 'F3-C3', 'C3-P3', 'P3-O1',
'FP2-F4', 'F4-C4', 'C4-P4', 'P4-O2',
'FP2-F8', 'F8-T8', 'T8-P8', 'P8-O2',
'FZ-CZ', 'CZ-PZ']
samplFreq=256 #sampling frequency of data
class SegSymbParams:
# WINDOW DISCRETIZATION
segLenSec=4 #length of discrete EEG windows on which to perform analysis
slidWindStepSec=0.5 #step of sliding window
labelVotingType='majority' #'majority', 'atLeastOne' or 'allOne' #defines how final label of a segment is chosen
# HD APPROACH
symbolType = 'StandardMLFeatures' #'CWT', 'Entropy', 'Amplitude', 'LBP', 'FFT','RawAmpl', 'StandardMLFeatures', 'AllFeatures'
numSegLevels = 20 #number levels on which to normalize values
# VARIOUS PARAMETERS OF APPROACHES
entropyType='spectral_entropy' #'perm_entropy', 'spectral_entropy', 'svd_entropy', 'app_entropy', 'sample_entropy'
#wavelet transform parameters
DWTfilterName='db4' #'sym5'
DWTlevel=0 # 0 means none, means automatic, 4,7,10, 0
CWTfilterName='cmor1.0-2.0' #'gaus1'
numFreqPerBand = 10 # only needed for CWT
CWTlevel=20 # only needed for CWT, the same as numSegLevels for others
noiseNormType = 'noiseNorm' #'noiseNorm', 'noNoiseNorm' defines if CWT values are normalized with CWT of noise
#for amplitude normalization min and mean values of whole signal are needed
amplitudeRangeFactor=2 # how much bigger and smaller values from mean we expect
amplitudeBinsSpacing='equal' #'equal','adjusted' #whether we make equal spacing bins between min and max values of we adjust them manually
minValueSignal=np.zeros((len(SigInfoParams.chToKeep))) #variable to keep track in real time about min and max value
meanValueSignal=np.zeros((len(SigInfoParams.chToKeep))) #variable to keep track in real time about min and max value
cntForAmplitude=0 #variable to keep track in real time about min and max value
# FREQUENCY BANDS - DEFINED FOR E.G. CWT
class EEGfreqBands:
freqVal4Lev = [2.5,6,11.5,22.5] #middle of alfa, beta, gamma, delta bands
freqVal10Lev = [0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5, 5]
freqVal20Lev = [0.25, 0.5, 0.75, 1, 1.25, 1.5, 1.75, 2, 2.25, 2.5, 2.75, 3, 3.25, 3.5, 4, 5, 6, 8, 10, 15]
freqVal30Lev = [0.1, 0.2, 0.4, 0.6, 0.8, 1.0, 1.2, 1.4, 1.6, 1.8, 2.0, 2.2, 2.4, 2.6, 2.8, 3.0, 3.4, 3.8, 4.2, 4.6,
5.0, 6.0, 7.0, 8.0, 10.0, 12.0, 14.0, 16.0, 18.0, 20.0]
class HDParams:
HDapproachON=1 # whether to use HD (1) or standardML (0)
HDvecType= 'bin' #'bin', 'bipol' #binary 0,1, bipolar -1,1
ItterativeRelearning='off' # 'on', 'off'
relearningImprovThresh=0.05
VotingType='' # '' for closest distance, 'ConfVoting'
# GENERAL PARAMETERS NEEDED FOR HD
CUDAdevice = 0 #number of the GPU used
D = 10000 #dimension of hypervectors
similarityType= 'hamming' #'hamming','cosine' #similarity measure used for comparing HD vectors
vectorTypeCh= 'random' # 'random','sandwich','scaleNoRand1','scaleNoRand2','scaleRand1', ,'scaleRand2' ... #defines how HD vectors are initialized
vectorTypeLevel = 'random'# 'random','sandwich','scaleNoRand1','scaleNoRand2','scaleRand1', ,'scaleRand2'... #defines how HD vectors are initialized
roundingTypeForHDVectors='inSteps' #'inSteps','onlyOne','noRounding' #defines how and when HD vectors are binarized
# LBP approach
LBPlen = 7 # 1 + dimension l of binary pattern
totalNumberBP = 2 ** (LBPlen - 1)
# more features
numFeat=45 #3 for similar to symbolization, 45 for standard Ml features
vectorTypeFeat='random'
bindingFeatures='ChxFeatxVal' #'FeatxVal', 'ChxFeatxVal', 'FeatxChxVal', 'ChFeatCombxVal' #defines how HD vectors encoded
normValuesForFeatures=np.zeros((2,numFeat)) #variable to keep track in real time
# FFT encoding
FFTUpperBound=16 # upped border for freq of interest, in Hz (up to 128, but should be mod2 of 128)
vectorTypeFreq='random'# 'random','sandwich','scaleNoRand1','scaleNoRand2','scaleRand1', ,'scaleRand2' #defines how HD vectors are initialized
bindingFFT = 'FreqxVal' # 'FreqxVal', 'ChxFreqxVal', 'PermChFreqxVal', 'FeatAppend' #defines how HD vectors encoded
#Raw amplitude encoding
bindingRawAmpl='ValxCh' #'ValxCh','PermValSamplxCh','PermValSampl' #defines how HD vectors encoded
class StandardMLParams:
modelType='DecisionTree' #'KNN', 'SVM', 'DecisionTree', 'RandomForest','BaggingClassifier','AdaBoost'
trainingDataResampling='NoResampling' #'NoResampling','ROS','RUS','TomekLinks','ClusterCentroids','SMOTE','SMOTEtomek'
samplingStrategy='default' # depends on resampling, but if 'default' then default for each resampling type, otherwise now implemented only for RUS if not default
#KNN parameters
KNN_n_neighbors=5
KNN_metric='euclidean' #'euclidean', 'manhattan'
#SVM parameters
SVM_kernel = 'linear' # 'linear', 'rbf','poly'
SVM_C = 1 # 1,100,1000
SVM_gamma = 'auto' # 0 # 0,10,100
#DecisionTree and random forest parameters
DecisionTree_criterion = 'gini' # 'gini', 'entropy'
DecisionTree_splitter = 'best' # 'best','random'
DecisionTree_max_depth = 0 # 0, 2, 5,10,20
RandomForest_n_estimators = 100 #10,50, 100,250
#Bagging, boosting classifier parameters
Bagging_base_estimator='SVM' #'SVM','KNN', 'DecisionTree'
Bagging_n_estimators = 100 # 10,50, 100,250
#creating features
class FeaturesParams:
numStandardFeat=45 #3, 45 number of features
featNorm='noNorm' #'Norm','noNorm'
winLen = 15 #nmber of last symbols to use for ML as features
typeCreatingFeatures='pastComposition' #'pastSequence','pastComposition'
numMostCorrelatedCh=9 #calculate (common) features only for n most correlated channels, use 2,4,9 to be able to have spatial and non spatial type
channelsAnalysisApproach='mostCorrelatedCh' # 'eachIndividualy','mostCorrelatedCh','groupedAllCh'
channelLabelsGrouping='allOne' #'majority' , 'atLeastOne' or 'allOne' #determining one label from labels of each channel
combiningChannelSequences= 'averageValue' #'mostCommmon', 'averageValue', 'combineToNewLabel'
combiningChannels= 'Spacial' #'Spacial', 'noSpacial'
combiningChMajorityThresh=0.25
combineMoreApproaches='appendFeatures' #'appendFeatures', 'createNewSymbols'
# class ZeroCrossFeatureParams:
# EPS_thresh = 64 #16, 32, 64, 128, 256
# buttFilt_order=4
# buttFilt_freq=20
# samplFreq=256
# minValueSignal = np.zeros((len(SigInfoParams.chToKeep)))
# meanValueSignal = np.zeros((len(SigInfoParams.chToKeep))) # variable to keep track in real time about min and max value
# cntForLL = 0 # variable to keep track in real time about min and max value
#SAVING SETUP once again to update if new info
with open('../PARAMETERS.pickle', 'wb') as f:
pickle.dump([GeneralParams, SegSymbParams, SigInfoParams, EEGfreqBands, StandardMLParams, FeaturesParams, patients], f)

Event Timeline