def normal2():
m = stocPy.normal(0, 1, obs=True)
v = stocPy.invGamma(3, 1)
cond = None
if init:
cond = obs
stocPy.normal(m, math.sqrt(v), cond)
def hmm():
states = []
states.append(stocPy.stocPrim("categorical", (sProbs,), obs=True, part=pind))
for i in range(1,17):
#print states
states.append(stocPy.stocPrim("categorical", (tProbs[states[i-1]],), obs=True, part=pind))
stocPy.normal(eMeans[states[i]], 1, cond=obs[i-1])
def normal9Dec15():
n = 15
var = 1.0
ms = []
for i in range(n):
ms.append(stocPy.normal(0, math.sqrt(var/(2**(i+1))), obs=True))
ms.append(stocPy.normal(0, math.sqrt(var/(2**n)), obs=True))
m = sum(ms)
def hmmExp():
states = []
states.append(categoricalExp(sProbs, name="s0"))
for i in range(1,17):
states.append(categoricalExp(tProbs[states[i-1]], name="s"+str(i)))
stocPy.normal(eMeans[states[i]], 1, cond=obs[i-1], name="c"+str(i))
global stateHist
global singleStateHist
curTime = time.time() - startTime
singleStateHist[curTime] = states[sind]
def normal8DecU20():
n = stocPy.stocPrim("randint", (0, 21))
var = 1.0
ms = []
for i in range(n):
ms.append(stocPy.normal(0, math.sqrt(var/(2**(i+1))), obs=True))
ms.append(stocPy.normal(0, math.sqrt(var/(2**n)), obs=True))
m = 10000 * ss.norm.cdf(sum(ms), loc = 0, scale = 1)
for datum in normalData:
stocPy.normal(m, 1, datum)
def normal7Dec15():
n = 15
var = 1.0
ms = []
for i in range(n):
ms.append(stocPy.normal(0, math.sqrt(var/(2**(i+1))), obs=True))
ms.append(stocPy.normal(0, math.sqrt(var/(2**n)), obs=True))
m = 100 * ss.norm.cdf(sum(ms), loc = 0, scale = 1)
for datum in normalData:
stocPy.normal(m, 1, datum)
def normal6Dec15():
n = 15
var = 10.0
ms = []
for i in range(n):
ms.append(stocPy.normal(0, math.sqrt(var/(2**(i+1))), obs=True))
ms.append(stocPy.normal(0, math.sqrt(var/(2**n)), obs=True))
m = sum(ms)
for datum in normalData:
stocPy.normal(m, 1, datum)
def hmm1():
states = []
states.append(categorical(sProbs))
for i in range(1,17):
states.append(categorical(tProbs[states[i-1]]))
for i in range(1,17):
stocPy.normal(eMeans[states[i]], 1, cond=obs[i-1])
global singleStateHist
curTime = time.time() - startTime
singleStateHist[curTime] = states[sind]
def dpmLazy():
crp = stocPy.crp(1.72)
sds = {}
ms = {}
for i in range(len(obs)):
c = crp(i)
if c not in ms:
sds[c] = math.sqrt(10 * stocPy.stocPrim("invgamma", (1, 0, 10), part=4))
ms[c] = stocPy.stocPrim("normal", (0, sds[c]), part=4)
stocPy.normal(ms[c], sds[c], obs[i])
obsLens.append(len(ms))
def normal4():
global ms
global vs
m = stocPy.normal(0, 1, obs=True)
if m > 0:
v = 1.0/3
else:
v = stocPy.invGamma(3, 1)
vs.append(v)
stocPy.normal(m, math.sqrt(v), cond=obs)
ms.append(m)
def dpmEager():
crp = stocPy.crp(1.72, 10)
sds = {}
ms = {}
cs = {}
for ps in range(len(crp)):
sds[ps] = math.sqrt(10 * stocPy.stocPrim("invgamma", (1, 0, 10), part=2))
ms[ps] = stocPy.stocPrim("normal", (0, sds[ps]), part=2)
for p in crp[ps]:
cs[p] = ps
for i in range(len(obs)):
stocPy.normal(ms[cs[i]], sds[cs[i]], obs[i])
obsLens.append(len(ms))
def tdfDecomp(init=False):
ys = []
d1 = stocPy.normal(0, 1, obs=init)
d2 = stocPy.normal(0, 2, obs=init)
d3 = stocPy.normal(0, 4, obs=init)
d4 = stocPy.normal(0, 8, obs=init)
d5 = stocPy.normal(0, 16, obs=init)
dof = 2 + 98 * ss.norm.cdf(d1+d2+d3+d4+d5, loc = 0, scale = math.sqrt(sum([x**2 for x in [1,2,4,8,16]])))
for i in range(len(conds)):
cond = None
if init:
cond = conds[i]
ys.append(stocPy.studentT(dof, cond=cond))
return ys
def normal3():
global ms
global vs
m = stocPy.normal(0, 1, obs=True)
if m > 0:
v = 1
else:
v = stocPy.invGamma(3, 1)
vs.append(v)
cond = None
if init:
cond = obs
stocPy.normal(m, math.sqrt(v), cond)
ms.append(m)
def hmmSpec():
states = []
states.append(categorical(sProbs))
for i in range(1,17):
states.append(categorical(tProbs[states[i-1]]))
stocPy.normal(eMeans[states[i]], 1, cond=obs[i-1])
#global singleStateHist
curTime = time.time() - startTime
#singleStateHist[curTime] = states[sind]
global stateHist
for i in range(len(states)):
try:
stateHist[i][curTime] = states[i]
except:
stateHist.append({curTime:states[i]})
def normal1Dec():
m1 = stocPy.normal(0, math.sqrt(0.5), obs=True)
m2 = stocPy.normal(0, math.sqrt(0.25), obs=True)
m3 = stocPy.normal(0, math.sqrt(0.125), obs=True)
m4 = stocPy.normal(0, math.sqrt(0.0625), obs=True)
m5 = stocPy.normal(0, math.sqrt(0.0625), obs=True)
m = m1+m2+m3+m4+m5
cond = None
if init:
cond = obs
stocPy.normal(m, 1, cond)
def normal2Dec1():
m1 = stocPy.normal(0, math.sqrt(0.5), obs=True)
m2 = stocPy.normal(0, math.sqrt(0.25), obs=True)
m3 = stocPy.normal(0, math.sqrt(0.125), obs=True)
m4 = stocPy.normal(0, math.sqrt(0.0625), obs=True)
m5 = stocPy.normal(0, math.sqrt(0.0625), obs=True)
m = m1+m2+m3+m4+m5
v = stocPy.invGamma(3, 1)
cond = None
if init:
cond = obs
stocPy.normal(m, math.sqrt(v), cond)
def normal4Dec1():
global ms
global vs
m1 = stocPy.normal(0, math.sqrt(0.5), obs=True)
m2 = stocPy.normal(0, math.sqrt(0.25), obs=True)
m3 = stocPy.normal(0, math.sqrt(0.125), obs=True)
m4 = stocPy.normal(0, math.sqrt(0.0625), obs=True)
m5 = stocPy.normal(0, math.sqrt(0.0625), obs=True)
m = m1+m2+m3+m4+m5s
if m > 0:
v = 1.0/3
else:
v = stocPy.invGamma(3, 1)
vs.append(v)
cond = None
if init:
cond = obs
stocPy.normal(m, math.sqrt(v), cond)
ms.append(m)
def normal8Old():
m = stocPy.unifCont(0, 10000, obs=True)
for datum in normalData:
stocPy.normal(m, 1, datum)
def normal1(): m = stocPy.normal(0, 1, obs=True) stocPy.normal(m, 1, obs)
def normalMean(): mean = stocPy.normal(1, math.sqrt(5), obs=True) stocPy.normal(mean, math.sqrt(2), cond=9) stocPy.normal(mean, math.sqrt(2), cond=8)
def marsagliaMean(): global sampleInd mean = marsaglia(1, 5) stocPy.normal(mean, math.sqrt(2), cond=9) stocPy.normal(mean, math.sqrt(2), cond=8) obsMean.append(mean)
def normal4Dec2():
global ms
global vs
m1 = stocPy.normal(0, math.sqrt(0.5), obs=True)
m2 = stocPy.normal(0, math.sqrt(0.25), obs=True)
m3 = stocPy.normal(0, math.sqrt(0.125), obs=True)
m4 = stocPy.normal(0, math.sqrt(0.0625), obs=True)
m5 = stocPy.normal(0, math.sqrt(0.0625), obs=True)
m = m1+m2+m3+m4+m5
if m > 0:
v = 1.0/3
else:
v1 = stocPy.normal(0, math.sqrt(0.5))
v2 = stocPy.normal(0, math.sqrt(0.25))
v3 = stocPy.normal(0, math.sqrt(0.125))
v4 = stocPy.normal(0, math.sqrt(0.0625))
v5 = stocPy.normal(0, math.sqrt(0.0625))
vn = ss.norm.cdf(v1+v2+v3+v4+v5, loc=0, scale=1)
v = ss.invgamma.ppf(vn, 3, loc=0, scale=1)
vs.append(v)
cond = None
if init:
cond = obs
stocPy.normal(m, math.sqrt(v), cond)
ms.append(m)
def normal2Dec2():
m1 = stocPy.normal(0, math.sqrt(0.5), obs=True)
m2 = stocPy.normal(0, math.sqrt(0.25), obs=True)
m3 = stocPy.normal(0, math.sqrt(0.125), obs=True)
m4 = stocPy.normal(0, math.sqrt(0.0625), obs=True)
m5 = stocPy.normal(0, math.sqrt(0.0625), obs=True)
m = m1+m2+m3+m4+m5
v1 = stocPy.normal(0, math.sqrt(0.5))
v2 = stocPy.normal(0, math.sqrt(0.25))
v3 = stocPy.normal(0, math.sqrt(0.125))
v4 = stocPy.normal(0, math.sqrt(0.0625))
v5 = stocPy.normal(0, math.sqrt(0.0625))
vn = ss.norm.cdf(v1+v2+v3+v4+v5, loc=0, scale=1)
v = ss.invgamma.ppf(vn, 3, loc=0, scale=1)
cond = None
if init:
cond = obs
stocPy.normal(m, math.sqrt(v), cond)
def normal9(): #prior = post m = stocPy.normal(0, 1, obs=True)
def normal8Part2():
m = stocPy.stocPrim("uniform", (0, 10000), obs=True, part=10)
for datum in normalData:
stocPy.normal(m, 1, datum)
def normal8Part():
m = stocPy.stocPrim("uniform", (0, 10000), obs=True, part=stocPy.stocPrim("randint", (0, 21)))
for datum in normalData:
stocPy.normal(m, 1, datum)
def normal8():
m = stocPy.stocPrim("uniform", (0, 10000), obs=True)
for datum in normalData:
stocPy.normal(m, 1, datum)
def normal6():
m = stocPy.normal(0, 10, obs=True)
for datum in normalData:
stocPy.normal(m, 1, datum)