def plotSkew(dgps):
global FII
cls=['$c_l$','$b$','$c_l$','$c_l$','$c_l$','$c_l$']
figure(size=3,aspect=0.6)
for h in range(len(dgps)):
try:D=dgps[h].loadData(i=2,j=2)
except FileNotFoundError: continue
x=dgps[h].cl
for g in range(3):
ax=subplot(3,len(dgps),1+h+g*len(dgps))
plt.grid(axis='x')
if g==0:
plt.title(DLBLS[h])
y=np.median(D[:,:,6],axis=1)*[1,0.1][dgps[h].suf=='BE']
elif g==1: y=np.median(D[:,:,7],axis=1)
elif g==2: y=np.median(D[:,:,23],axis=1)
plt.plot(x,y,CLRS[-1],lw=LWS[1])
if g==0: plt.ylim([0,1])
elif g==1: plt.ylim([[0,10],[0,0.01],[0,3],[0,6],[0,0.1],[0,0.06]][h])
elif g==2: plt.ylim([[0,3],[0,3],[0,8],[0,3],[0,3],[-3,3]][h])
if g<2: ax.set_xticklabels([])
else: plt.xlabel(cls[h])
plt.xlim([x[0],x[-1]])
ax.locator_params(axis='x', nbins=4,min_n_ticks=3)
if not h: plt.ylabel(['Mean','Variance','Skewness'][g])
elif g==0: ax.set_yticklabels([])
elif g==1:
lm=[0,-3,0,0,-2,-2][h]
if lm!=0: ax.ticklabel_format(axis='y',style='sci',scilimits=(lm,lm))
if FII>=0:
fn='fig%02d.%s'%(FII,FIGFMT)
FII+=1
else: fn=FIGLBLS[0]+'png'
plt.savefig(FIGDIR+fn,bbox_inches='tight',dpi=DPI,format=FIGFMT)
plt.clf();plt.close()
def plotPDF(A,B,C,pdf,xmax=1,albls=['','','test']):
C=np.linspace(C[0],C[-1],6)
x=np.linspace(0,xmax,101)
figure(size=3,aspect=1)
def _hlp(A,i):
if A.size>1 and xmax>0: return '%.2f'%A[i]
else: return ''
for i in range(A.size):
for j in range(B.size):
for h in range(C.size):
ax=subplot(A.size,B.size,B.size*i+1+j)
try:
if xmax>0: y=pdf(x,A[i],B[j],C[h])
else: y,x=pdf(A[i],B[j],C[h])
except AssertionError: continue
plt.plot(x,y,'k',alpha=0.2)
plt.ylim([0,1])
ax.set_yticks([0,1])
if j>0: ax.set_yticklabels([])
else: plt.ylabel(albls[0]+_hlp(A,i))
if i+1<A.size: ax.set_xticklabels([])
else: plt.xlabel(albls[1]+_hlp(B,j))
plt.grid(False)
plt.savefig(SUPDIR+albls[-1]+'prob',format=FIGFMT,
bbox_inches='tight',dpi=DPI)
plt.clf();plt.close()
def plotBrief(x,R,y=np.nan,pref='', ylim=[[],[]]):
global FII
figure(size=2,aspect=1.2)
for f in range(1,7):
ax=subplot(3,2,f);
plt.grid(axis='x')
plt.ylim([-0.02,1.02])
plt.title(['CI','Cohen\'s $d$','TOST','Bayesian $t$ test','Two-group tests','Three-group tests'][f-1])
if f==1:
plotCI(x,R)
#if np.any(~np.isnan(y)): plt.plot(x,y,'c')
plt.ylim(ylim[0])
elif f==2:
plotCohendCI(x,R)
plt.ylim(ylim[1])
elif f==3: plotTOST(x,R[:,:,12:16])
elif f==4: plotBFT(x,R[:,:,16:18])
elif f==5:
plotNHST(x,R[:,:,8:12])
elif f==6: plotANOVA1F(x,R[:,:,18:21])
if f<5: ax.set_xticklabels([])
else: plt.xlabel(['$c_l$','$b$'][int(pref=='W')])
plt.xlim([x[0],x[-1]])
if FII>=0:
fn='fig%02d.%s'%(FII,FIGFMT)
FII+=1
else: fn=pref+'brief.png'
plt.savefig(FIGDIR+fn,dpi=DPI,bbox_inches='tight',format=FIGFMT)
def plot(figname='stan.png', dpi=300):
from matusplotlib import figure, subplot, plt
figure(size=3, aspect=0.3)
il = [
'dog', 'trolley', 'wallet', 'plane', 'resume', 'kitten', 'mean score',
'median score'
]
w = loadStanFit('schnall')
S = np.load('S.npy')
offset = np.array(w['c'][:, 0] / 2 + w['c'][:, -1] / 2, ndmin=2).T
scale = np.array(np.abs(w['c'][:, 0] - w['c'][:, -1]) / 2, ndmin=2).T
bp = np.linspace(-2.2, 2.2, 51)
b = bp * np.median(scale) + np.median(offset)
cs = np.median(w['c'], axis=0)
d = np.median(-w['d'], axis=0)
#xlm=[cs[0]-0.2,cs[-1]+0.2]
xlm = b[[0, -1]]
cls = []
for i in range(cs.size):
cls.append('$c_%d$' % i)
tmp = np.median(-w['tbeta'], axis=0)
for j in range(2):
ax = subplot(1, 2, 1 + j)
plt.plot(xlm, [-d[0], -d[0]])
plt.xlim(xlm)
ax.set_xticks(cs)
ax.set_xticklabels(cls)
plt.plot(tmp[0, :], -0.7 * np.ones(6), 'xg')
plt.plot(tmp[1, :], -0.9 * np.ones(6), 'xr')
#for k in range(tmp.shape[1]):
# plt.plot(tmp[0:,k],[-0.09,-0.11],'k',alpha=0.2)
ds = np.zeros((len(S), 3)) * np.nan
for i in range(len(S)):
if j:
wi = loadStanFit('schnSimBB%02d' % i)
ds[i, :] = sap(wi['gg1'][:, 1] - wi['gg2'][:, 1],
[2.5, 50, 97.5])
elif j == 0:
wi = loadStanFit('schnSimOLR%02d' % i)
ds[i, :] = sap(wi['d'], [2.5, 50, 97.5])
plt.plot(b, ds[:, 1], 'k')
temp = [list(b) + list(b)[::-1], list(ds[:, 0]) + list(ds[:, 2])[::-1]]
ax.add_patch(
plt.Polygon(xy=np.array(temp).T, alpha=0.2, fc='k', ec='k'))
plt.ylim([[-2, 4], [-2, 4]][j])
plt.ylabel('$c_u^\Delta$')
plt.xlabel(['$c_u=-c_l$', '$c_u$'][j])
plt.title(['OLRM', 'Beta-Binomial'][j])
plt.grid(axis='x')
plt.savefig(figname, bbox_inches='tight', dpi=dpi)
def makeSupplement(self):
print(self.suf+'creating supplement')
#plotPDF(self.np1,self.np2,self.cl,self.pdf,xmax=self.xmax,
# albls=['$c_l^\Delta=$','$%s=$'%self.np2lbl,self.suf])
#tref=(self.cld[0,1]-self.cld[0,0])*np.ones(self.cl.shape)
n1=self.np1.size;n2=self.np2.size
for f in range(1,7):
plt.close('all')
figure(num=f+1,size=3,aspect=1)
for i in range(n1):
for j in range(n2):
R=self.loadData(i=i,j=j)
ax=subplot(n1,n2,i*n2+j+1)
#plt.title(getTitle(i,j))
plt.grid(axis='x')
plt.ylim([-0.05,1.05])
x=np.load(DPATH+'cl_'+self.suf+'.npy')
legon= (i==0 and j==0)
if f==1:
plotCI(x,R)
tref=(self.cld[0,1]-self.cld[0,0])*self.np1[i]
plt.plot(x,tref*np.ones(x.size),'c')
plt.ylim(self.cil[i])
elif f==2:
plotCohendCI(x,R,ylim=self.cil[i+5])
elif f==3:plotNHST(x,R[:,:,8:12],legon)
elif f==4:plotANOVA1F(x,R[:,:,18:22],legon)
elif f==5:plotTOST(x,R[:,:,12:16],legon)
elif f==6:plotBFT(x,R[:,:,16:18],legon)
plt.xlim([x[0],x[-1]])
if j!=0:ax.set_yticklabels([])
else:plt.ylabel('$c_l^\Delta=%.2f$'%self.np1[i])
if i!=n1-1: ax.set_xticklabels([])
else:
temp=(self.np2lbl,self.np2[j])
plt.xlabel('$%s=%.2f$'%temp)
plt.savefig(SUPDIR+self.suf+FIGLBLS[f],
dpi=DPI,bbox_inches='tight',format=FIGFMT)
plt.clf();plt.close()
def plotANOVAall(dgps,showLegend=False,alpha=0.05,i=2,j=2):
global FII
cls=['$c_l$','$b$','$c_l$','$c_l$','$c_l$','$c_l$']
lblls=['no X','no ME','uncrossed','crossed',
'double-crossed','no X','uncrossed']
K=len(S)-4
figure(size=3,aspect=1)
for h in range(len(dgps)):
D=dgps[h].loadData(i=i,j=j)[:,:,AI:]
x=dgps[h].cl
for k in range(K):
ax=subplot(K,len(dgps),1+k*len(dgps)+h)
plt.grid(axis='x')
for ii in range(3):
s=np.all(~np.isnan(D[:,:,k*9]),axis=1)
if not k: plt.title(DLBLS[h])
plt.plot(x[s],(D[s,:,k*9+ii*3+2]<alpha).mean(1),
CLRS[ii],lw=LWS[ii],alpha=0.7)
if k<5:
plt.plot(x[s],(D[s,:,k*9+ii*3+1]<alpha).mean(1),
CLRS[ii]+'--',alpha=0.7,lw=LWS[ii])
plt.plot(x[s],(D[s,:,k*9+ii*3]<alpha).mean(1),
CLRS[ii]+':',alpha=0.7,lw=LWS[ii])
if not h: plt.ylabel(lblls[k])
else:ax.set_yticklabels([])
if k<K-1:ax.set_xticklabels([])
else: plt.xlabel(cls[h])
plt.ylim([-0.05,1.05])
plt.xlim([x[0],x[-1]])
plt.locator_params(axis='x', nbins=4)#,min_n_ticks=3)
if FII>=0:
fn='fig%02d.%s'%(FII,FIGFMT)
FII+=1
else: fn='ANOVA%d%d.png'%(i,j)
plt.savefig(FIGDIR+fn,bbox_inches='tight',dpi=DPI,format=FIGFMT)