def uniform_doseless(start, end, height=None):
scaled = False
if height is None:
height = 1 / (end - start)
scaled = True
conditional = SplineModelConditional()
conditional.add_segment(SplineModel.neginf, start, [SplineModel.neginf])
conditional.add_segment(start, end, [math.log(height)])
conditional.add_segment(end, SplineModel.posinf, [SplineModel.neginf])
return SplineModel(True, [''], [conditional], scaled)
def apply_as_distribution(self, model):
if model.kind() == 'ddp_model':
if len(model.xx) != len(self.yy):
raise ValueError("Wrong number of values.")
for ii in range(len(model.xx)):
if model.get_xx()[ii] != self.get_yy()[ii]:
raise ValueError("Wrong values in distribution.")
vv = self.lin_p().dot(model.lin_p())
return DDPModel('ddp1',
'distribution',
self.xx_is_categorical,
self.get_xx(),
model.yy_is_categorical,
model.get_yy(),
vv,
scaled=self.scaled)
elif model.kind() == 'spline_model':
if len(model.xx) != len(self.yy):
raise ValueError("Wrong number of values.")
for ii in range(len(model.xx)):
if model.get_xx()[ii] not in self.get_yy():
raise ValueError("Wrong values in distribution.")
pp = self.lin_p()
conditionals = []
for ii in range(len(self.xx)):
conds = []
for jj in range(len(model.xx)):
original = model.get_conditional(model.get_xx()[jj])
conditional = SplineModelConditional(
original.y0s, original.y1s, original.coeffs)
conditional.scale(
pp[ii, self.get_yy().index(model.get_xx()[jj])])
conds.append(conditional)
conditionals.append(
SplineModelConditional.approximate_sum(conds))
return SplineModel(self.xx_is_categorical,
self.get_xx(),
conditionals,
scaled=self.scaled)
else:
raise ValueError("Unknown model type in apply_as_distribution")
def test_exponential():
fp = StringIO.StringIO()
fp.write("dpc1,mean\n")
fp.write("0,1\n")
fp.seek(0)
output = StringIO.StringIO()
FeaturesInterpreter.init_from_feature_file(SplineModel(), fp, ',',
(0, 100)).write(output, ',')
print output.getvalue()
fp.close()
output.close()
def test_gtc(modelfile, distfile):
with open(modelfile, "r") as modelfp:
model = SplineModel()
FeaturesInterpreter.init_from_feature_file(model, modelfp, ',', (SplineModel.neginf, SplineModel.posinf))
with open(distfile, "r") as distfp:
dist = DistributionModel()
dist.init_from(distfp, ',')
output = StringIO.StringIO()
dist.apply_as_distribution(model).write(output, ',')
print output.getvalue()
output.close()
def test_non_gaussian():
fp = StringIO.StringIO()
fp.write("dpc1,mean,.025,.975\n")
fp.write("0,0,-1,2\n")
fp.seek(0)
output = StringIO.StringIO()
FeaturesInterpreter.init_from_feature_file(SplineModel(), fp, ',',
('-inf', 'inf')).write(
output, ',')
print output.getvalue()
fp.close()
output.close()
def polynomial(lowbound, highbound, betas, covas, num=40):
betas = np.array(betas)
covas = np.mat(covas)
if covas.shape[0] != covas.shape[1] and len(betas) != covas.shape[0]:
return "Error: Please provide a complete covariance matrix."
if np.any(linalg.eig(covas)[0] < 0):
return "Error: Covariance matrix is not positive definite."
xx = np.linspace(lowbound, highbound, num=num)
xxs = {}
for x in xx:
xvec = np.mat([[1, x, x**2, x**3][0:len(betas)]])
serr = np.sqrt(xvec * covas * np.transpose(xvec))
xxs[x] = (betas.dot(np.squeeze(np.asarray(xvec))), serr[0,0])
return SplineModel.create_gaussian(xxs, xx, False)
import sys, StringIO
sys.path.append("../lib/models")
from model import Model
from ddp_model import DDPModel
from spline_model import SplineModel
mod1 = DDPModel.create_lin([0, 1], dict(control=[.5, .5], treatment=[0, 1]))
mod2 = DDPModel.create_lin([0, 1], dict(control=[.5, .5], treatment=[1, 0]))
def printmod(mod):
output = StringIO.StringIO()
mod.write(output, ',')
print output.getvalue()
printmod(Model.combine([mod1, mod2], [.5, .5]))
printmod(Model.combine([mod1, mod2], [.25, 2]))
mod1 = SplineModel.create_gaussian(dict(control=(0, 1), treatment=(1, 1e-3)))
mod2 = SplineModel.create_gaussian(dict(control=(0, 1), treatment=(0, 1e-3)))
def printmod2(mod):
output = StringIO.StringIO()
mod.write_gaussian_plus(output, ',')
print output.getvalue()
printmod2(Model.combine([mod1, mod2], [.5, .5]))
printmod2(Model.combine([mod1, mod2], [.25, 2]))
from model import Model
from ddp_model import DDPModel
from spline_model import SplineModel
mod1 = DDPModel.create_lin([0, 1], dict(control=[.5, .5], treatment=[0, 1]))
mod2 = DDPModel.create_lin([0, 1], dict(control=[.5, .5], treatment=[1, 0]))
def printmod(mod):
output = StringIO.StringIO()
mod.write(output, ',')
print output.getvalue()
printmod(Model.combine([mod1, mod2], [.5, .5]))
printmod(Model.combine([mod1, mod2], [.25, 2]))
mod1 = SplineModel.create_gaussian(dict(control=(0, 1), treatment=(1, 1e-3)))
mod2 = SplineModel.create_gaussian(dict(control=(0, 1), treatment=(0, 1e-3)))
def printmod2(mod):
output = StringIO.StringIO()
mod.write_gaussian_plus(output, ',')
print output.getvalue()
printmod2(Model.combine([mod1, mod2], [.5, .5]))
printmod2(Model.combine([mod1, mod2], [.25, 2]))