

In [1]:

    
using Revise,FileIO
y1,x1,y2,x2=load("dirori.jld","y1","x1","y2","x2")
rx1=deg2rad.(x1);rx2=deg2rad.(x2);

Direction and Orientation Tuning Curve



In [3]:

    
using NeuroAnalysis,DataFrames
plotcondresponse(y1,DataFrame(Orientation=x1),:Orientation)









    Out[3]:



In [4]:

    
plotcondresponse(y1,DataFrame(Orientation=x1),:Orientation,projection=:polar)









    Out[4]:



In [5]:

    
plotcondresponse(y2,DataFrame(Orientation=x2),:Orientation)









    Out[5]:



In [6]:

    
plotcondresponse(y2,DataFrame(Orientation=x2),:Orientation,projection=:polar)









    Out[6]:

Circular Variance



In [7]:

    
d1=deg2rad(filter(x->x!=0,unique(diff(sort(x1))))[1]);d2=deg2rad(filter(x->x!=0,unique(diff(sort(x2))))[1])
arx1=circaxial.(rx1);arx2=circaxial.(rx2)
ad1=circaxial(d1);ad2=circaxial(d2);



In [8]:

    
DataFrame(dirvar1 = circvar(rx1,y1,d1),dirvar2 = circvar(rx2,y2,d2),orivar1 = circvar(arx1,y1,ad1),orivar2 = circvar(arx2,y2,ad2))









    Out[8]:




1 rows × 4 columns
dirvar1 dirvar2 orivar1 orivar2
Float64 Float64 Float64 Float64
1 0.230787 0.880278 0.3012 0.732038

von Mises and Generalized von Mises



In [9]:

    
using Plots
plot([vmf,gvmf],-pi,pi,label=["vmf","gvmf"])









    Out[9]:

GvM Model Fitting



In [10]:

    
using LsqFit
gvmfit1 = curve_fit((x,p)->gvmf.(x,p...),rx1,y1,Float64[1,0,0,0,0])
plot(x->gvmf(x,gvmfit1.param...),0,2pi)









    Out[10]:



In [11]:

    
plot(x->gvmf(x,gvmfit1.param...),0:0.01:2pi,proj=:polar)









    Out[11]:



In [12]:

    
gvmfit2 = curve_fit((x,p)->gvmf.(x,p...),rx2,y2,Float64[1,0,0,0,0])
plot(x->gvmf(x,gvmfit2.param...),0,2pi)









    Out[12]:



In [13]:

    
plot(x->gvmf(x,gvmfit2.param...),0:0.01:2pi,proj=:polar)









    Out[13]:

vM Model Fitting



In [14]:

    
vmfit1 = curve_fit((x,p)->vmf.(x,p...,n=2),rx1,y1,Float64[1,0,0])
plot(x->vmf(x,vmfit1.param...,n=2),0,2pi)









    Out[14]:



In [15]:

    
plot(x->vmf(x,vmfit1.param...,n=2),0:0.01:2pi,proj=:polar)









    Out[15]:



In [16]:

    
vmfit2 = curve_fit((x,p)->vmf.(x,p...,n=2),rx2,y2,Float64[1,0,0])
plot(x->vmf(x,vmfit2.param...,n=2),0,2pi)









    Out[16]:



In [17]:

    
plot(x->vmf(x,vmfit2.param...,n=2),0:0.01:2pi,proj=:polar)









    Out[17]:



In [ ]: