ipyPaste.py

    import scipy as sp
    import numpy as np
    import nibabel as nb
    import scipy.signal as sig
    import matplotlib.pyplot as pl
    from multiprocessing import Pool
    import multiprocessing
    import time


    def MakeT(Int,Start,NumSlices,Tr):
        dt=float(Tr)/float(NumSlices)
        IntSeq=[]
    
        for i in range(Int):
            IntSeq=IntSeq+range(i,NumSlices,Int)
        
        IntSeq=np.array(IntSeq)
        IntSeq=np.roll(IntSeq,Start,0)
        IntTime=IntSeq*dt
    
        return(IntSeq,IntTime)


    def bandpass(CutOffFreq, SamplingRate, StopGain, TranWidth):
        NiquistRate = SamplingRate/2.0
        N, beta = sig.kaiserord(StopGain,TranWidth/NiquistRate)
        print 'the order of the FIR filter is:' + str(N) + ', If this is bigger than the size of the data please adjust the width and gain of the filter'
        taps = sig.firwin(N, CutOffFreq, window=('kaiser', beta), pass_zero=False, scale=True, nyq=NiquistRate)

        return (taps,NiquistRate)



    def loadNIIimg(Fname):
        img=nb.load(Fname).get_data()
        return img


    Tr = 2.0
    Int = 6
    Foriginal = 0.5 # Hz
    Fnew = 20.0 #Hz
    Stopgain = 60
    Tranwidth = 0.08
    print("Designing Filter")    
    BPF, Nyq = bandpass([0.01, 0.21], Fnew, Stopgain, Tranwidth)
    Z=4
    tShift=0.5

    Ny=45

    #print('SmallSig')
    print("STarting Slice "+ str(Z))
    #Sig2=np.squeeze(Im[Ny,Ny,Z,:])

    Sig=np.loadtxt('/home/dparker/Desktop/Signal.txt')
    tdim=len(list(np.shape(Sig)))-1
    


    tShift=-1
    LTS=Sig.shape[-1]
    FR=np.array(range(int(round(LTS/2.)),0,-1))
    BR=np.array(range(LTS-2,int(FR.shape[-1]-1)-2,-1))


    if tdim==0:
        
        FrontPad=(Sig[FR])	
        BackPad=Sig[BR]
        LFP=FrontPad.shape[-1]
    elif tdim==1:
        
        FrontPad=Sig[:,FR]
        BackPad=Sig[:,BR]
        

        LFP=FrontPad.shape[-1]

    elif tdim==2:
        
        FrontPad=Sig[:,:,FR]
        BackPad=Sig[:,:,BR]

        LFP=FrontPad.shape[-1]

    elif tdim==3:
        
        FrontPad=Sig[:,:,:,FR]
        BackPad=Sig[:,:,:,BR]

        LFP=FrontPad.shape[-1]

    else:
        print('Bad Array Dimensions for Padding')

    Sig=np.concatenate((FrontPad,Sig,BackPad),-1)
    #print('ZeroPadding')
    
    S=int(round(Fnew/Foriginal))
    Dm=list(np.shape(Sig))
    SS=Dm
    Dm[tdim]=Dm[tdim]*S


    ZeroSig=np.zeros(Dm)

    k=0
    if tdim==0:
        ZeroSig[::S]=Sig
    elif tdim==1:       
        ZeroSig[:,::S]=Sig
    elif tdim==2:        
        ZeroSig[:,:,::S]=Sig                 
    elif tdim==3:       
        ZeroSig[:,:,:,::S]=Sig
    else:
        print("Bad Array Dimensions")  
    
    #print('Padding Finished')

    del Sig    
    print ZeroSig.shape[-1]
    #print('Filtering')
    ZeroSig = sig.filtfilt(BPF, [1], ZeroSig*float(S), axis=-1,padtype='even', padlen=0) 
    #print('Filtering Finished')


    #print('Shifting')

    tdim=len(SS)-1
    Fs=Fnew
    Ts=1/Fs

    Sig=np.zeros(SS)
    shift=round(tShift*Fs)
    print shift

    if tdim==0:
    
        if shift>0:
            print ZeroSig.shape
            Rep=np.tile(ZeroSig[0],shift)
            ZeroSig=np.append(Rep,ZeroSig[0:-shift],-1)
            print ZeroSig.shape
            Sig=ZeroSig[range(0,ZeroSig.shape[-1]-S,S)]       
        
        else:
            Rep=np.tile(ZeroSig[-1],abs(shift))
            ZeroSig=np.append(ZeroSig,Rep,-1)
            Sig=ZeroSig[range(int(abs(shift)),ZeroSig.shape[-1]-1,S)]
            
        Sig=Sig[LFP:LFP+LTS]
        
    elif tdim==1:  
     
        if shift>0:
            Rep=np.tile(np.expand_dims(ZeroSig[:,0],axis=-1),[1,shift])
            ZeroSig=np.append(Rep,ZeroSig,-1)
            Sig=ZeroSig[:,range(0,ZeroSig.shape[-1]-S,S)]       
    
        else:
            Rep=p.tile(np.expand_dims(ZeroSig,axis=-1),[1,abs(shift)])
            ZeroSig=np.append(ZeroSig,Rep,-1)
            Sig=ZeroSig[:,range(int(abs(shift)),ZeroSig.shape[-1]-1,S)]
            
        Sig=Sig[:,LFP:LFP+LTS-1]
  
    elif tdim==2: 

        if shift>0:
    
            Rep=np.tile(np.expand_dims(ZeroSig[:,:,0],axis=-1),[1,1,shift])
            ZeroSig=np.append(Rep,ZeroSig,-1)
            Sig=ZeroSig[:,:,range(0,ZeroSig.shape[-1]-S,S)]       
    
        else:
            Rep=np.tile(np.expand_dims(ZeroSig[:,:,-1],axis=-1),[1,1,abs(shift)])
            ZeroSig=np.append(ZeroSig,Rep,-1)
            Sig=ZeroSig[:,:,range(int(abs(shift)),ZeroSig.shape[-1]-1,S)]
        
        Sig=Sig[:,:,LFP:LFP+LTS-1]

       
    elif tdim==3:   
        if shift>0:
    
            Rep=np.tile(np.expand_dims(ZeroSig[:,:,:,0],axis=-1),[1,1,1,shift])
            Rep=np.expand_dims(Rep,axis=-1)
            ZeroSig=np.append(Rep,ZeroSig,-1)
            Sig=ZeroSig[:,:,:,range(0,ZeroSig.shape[-1]-S,S)]       
    
        else:
            Rep=np.tile(np.expand_dims(ZeroSig[:,:,:,-1],axis=-1),[1,1,1,shift])

            Rep=np.expand_dims(Rep,axis=-1)
            ZeroSig=np.append(ZeroSig,Rep,-1)
            Sig=ZeroSig[:,:,:,range(int(abs(shift)),ZeroSig.shape[-1]-1,S)]
        
        Sig=Sig[:,:,:,LFP:LFP+LTS-1]

    else:       
        print("Bad Array Dimensions")  

    Sig2=np.loadtxt('/home/dparker/Desktop/Signal.txt')
    t=np.array(range(201))*2
    pl.plot(t,Sig)
    pl.plot(t,Sig2)
    pl.show()