def test_lumi(backend):
mc = MockConfig(
par_map={
'lumi': {
'paramset': constrained_by_normal(
n_parameters=1,
inits=[0],
bounds=[[-5, 5]],
fixed=False,
auxdata=[None],
sigmas=[None],
),
'slice': slice(0, 1),
}
},
par_order=['lumi'],
samples=['signal', 'background'],
)
mega_mods = {
'lumi/lumi': {
'signal': {
'type': 'lumi',
'name': 'lumi',
'data': {'mask': [True, True, True]},
},
'background': {
'type': 'lumi',
'name': 'lumi',
'data': {'mask': [True, True, True]},
},
},
}
hsc = lumi_combined([('lumi', 'lumi')], mc, mega_mods)
mod = hsc.apply(pyhf.tensorlib.astensor([0.5]))
shape = pyhf.tensorlib.shape(mod)
assert shape == (1, 2, 1, 3)
mod = np.asarray(pyhf.tensorlib.tolist(mod))
assert np.allclose(mod[0, 0, 0], [0.5, 0.5, 0.5])
assert np.allclose(mod[0, 1, 0], [0.5, 0.5, 0.5])
hsc = lumi_combined([('lumi', 'lumi')], mc, mega_mods, batch_size=4)
mod = hsc.apply(pyhf.tensorlib.astensor([[1.0], [2.0], [3.0], [4.0]]))
shape = pyhf.tensorlib.shape(mod)
assert shape == (1, 2, 4, 3)
mod = np.asarray(pyhf.tensorlib.tolist(mod))
assert np.allclose(mod[0, 0, 0], [1.0, 1.0, 1.0])
assert np.allclose(mod[0, 0, 1], [2.0, 2.0, 2.0])
assert np.allclose(mod[0, 0, 2], [3.0, 3.0, 3.0])
assert np.allclose(mod[0, 0, 3], [4.0, 4.0, 4.0])
def test_histosys(backend):
mc = MockConfig(
par_map={
'hello': {
'paramset':
constrained_by_normal(
name='hello',
is_scalar=True,
n_parameters=1,
inits=[0],
bounds=[[-5, 5]],
fixed=False,
auxdata=[0.0],
),
'slice':
slice(0, 1),
},
'world': {
'paramset':
constrained_by_normal(
name='world',
is_scalar=True,
n_parameters=1,
inits=[0],
bounds=[[-5, 5]],
fixed=False,
auxdata=[0.0],
),
'slice':
slice(1, 2),
},
},
par_order=['hello', 'world'],
samples=['signal', 'background'],
)
mega_mods = {
'histosys/hello': {
'signal': {
'type': 'histosys',
'name': 'hello',
'data': {
'hi_data': [11, 12, 13],
'lo_data': [9, 8, 7],
'nom_data': [10, 10, 10],
'mask': [True, True, True],
},
},
'background': {
'type': 'histosys',
'name': 'hello',
'data': {
'hi_data': [11, 12, 13],
'lo_data': [9, 8, 7],
'nom_data': [10, 10, 10],
'mask': [True, True, True],
},
},
},
'histosys/world': {
'signal': {
'type': 'histosys',
'name': 'world',
'data': {
'hi_data': [10, 10, 10],
'lo_data': [5, 6, 7],
'nom_data': [10, 10, 10],
'mask': [True, True, True],
},
},
'background': {
'type': 'histosys',
'name': 'world',
'data': {
'hi_data': [10, 10, 10],
'lo_data': [5, 6, 7],
'nom_data': [10, 10, 10],
'mask': [True, True, True],
},
},
},
}
hsc = histosys_combined([('hello', 'histosys'), ('world', 'histosys')], mc,
mega_mods)
mod = hsc.apply(pyhf.tensorlib.astensor([0.5, -1.0]))
shape = pyhf.tensorlib.shape(mod)
assert shape == (2, 2, 1, 3)
mod = np.asarray(pyhf.tensorlib.tolist(mod))
assert np.allclose(mod[0, 0, 0], [0.5, 1.0, 1.5])
hsc = histosys_combined([('hello', 'histosys'), ('world', 'histosys')],
mc,
mega_mods,
batch_size=4)
mod = hsc.apply(
pyhf.tensorlib.astensor([[-1.0, -1.0], [1.0, 1.0], [-1.0, 1.0],
[1.0, 1.0]]))
shape = pyhf.tensorlib.shape(mod)
assert shape == (2, 2, 4, 3)
mod = np.asarray(pyhf.tensorlib.tolist(mod))
assert np.allclose(mod[0, 0, 0], [-1.0, -2.0, -3.0])
assert np.allclose(mod[0, 0, 1], [1.0, 2.0, 3.0])
assert np.allclose(mod[0, 0, 2], [-1.0, -2.0, -3.0])
assert np.allclose(mod[0, 0, 3], [1.0, 2.0, 3.0])
def test_stat(backend):
mc = MockConfig(
par_map={
'staterror_chan1': {
'paramset':
constrained_by_normal(
name='staterror_chan1',
is_scalar=False,
n_parameters=1,
inits=[1],
bounds=[[0, 10]],
fixed=False,
auxdata=[1],
),
'slice':
slice(0, 1),
},
'staterror_chan2': {
'paramset':
constrained_by_normal(
name='staterror_chan2',
is_scalar=False,
n_parameters=2,
inits=[1, 1],
bounds=[[0, 10], [0, 10]],
fixed=False,
auxdata=[1, 1],
),
'slice':
slice(1, 3),
},
},
channels=['chan1', 'chan2'],
channel_nbins={
'chan1': 1,
'chan2': 2
},
par_order=['staterror_chan1', 'staterror_chan2'],
samples=['signal', 'background'],
)
mega_mods = {
'staterror/staterror_chan1': {
'signal': {
'type': 'staterror',
'name': 'staterror_chan1',
'data': {
'mask': [True, False, False],
'nom_data': [10, 10, 10],
'uncrt': [1, 0, 0],
},
},
'background': {
'type': 'staterror',
'name': 'staterror_chan1',
'data': {
'mask': [True, False, False],
'nom_data': [10, 10, 10],
'uncrt': [1, 0, 0],
},
},
},
'staterror/staterror_chan2': {
'signal': {
'type': 'staterror',
'name': 'staterror_chan2',
'data': {
'mask': [False, True, True],
'nom_data': [10, 10, 10],
'uncrt': [0, 1, 1],
},
},
'background': {
'type': 'staterror',
'name': 'staterror_chan2',
'data': {
'mask': [False, True, True],
'nom_data': [10, 10, 10],
'uncrt': [0, 1, 1],
},
},
},
}
hsc = staterror_combined(
[('staterror_chan1', 'staterror'), ('staterror_chan2', 'staterror')],
mc,
mega_mods,
)
mod = hsc.apply(pyhf.tensorlib.astensor([1.1, 1.2, 1.3]))
shape = pyhf.tensorlib.shape(mod)
assert shape == (2, 2, 1, 3)
mod = np.asarray(pyhf.tensorlib.tolist(mod))
assert np.allclose(mod[0, 0, 0], [1.1, 1.0, 1.0])
assert np.allclose(mod[1, 0, 0], [1, 1.2, 1.3])
def test_batched_constraints(backend):
config = MockConfig(
par_order=['pois1', 'pois2', 'norm1', 'norm2'],
par_map={
'pois1': {
'paramset':
constrained_by_poisson(
name='pois1',
is_scalar=False,
n_parameters=1,
inits=[1.0],
bounds=[[0, 10]],
auxdata=[12],
factors=[12],
fixed=False,
),
'slice':
slice(0, 1),
'auxdata': [1],
},
'pois2': {
'paramset':
constrained_by_poisson(
name='pois2',
is_scalar=False,
n_parameters=2,
inits=[1.0] * 2,
bounds=[[0, 10]] * 2,
auxdata=[13, 14],
factors=[13, 14],
fixed=False,
),
'slice':
slice(1, 3),
},
'norm1': {
'paramset':
constrained_by_normal(
name='norm1',
is_scalar=False,
n_parameters=2,
inits=[0] * 2,
bounds=[[0, 10]] * 2,
auxdata=[0, 0],
sigmas=[1.5, 2.0],
fixed=False,
),
'slice':
slice(3, 5),
},
'norm2': {
'paramset':
constrained_by_normal(
name='norm2',
is_scalar=False,
n_parameters=3,
inits=[0] * 3,
bounds=[[0, 10]] * 3,
auxdata=[0, 0, 0],
fixed=False,
),
'slice':
slice(5, 8),
},
},
)
suggested_pars = [1.0] * 3 + [0.0] * 5 # 2 pois 5 norm
constraint = poisson_constraint_combined(config)
result = default_backend.astensor(
pyhf.tensorlib.tolist(
constraint.logpdf(
pyhf.tensorlib.astensor(config.auxdata),
pyhf.tensorlib.astensor(suggested_pars),
)))
assert np.isclose(
result,
sum([
default_backend.poisson_logpdf(data, rate)
for data, rate in zip([12, 13, 14], [12, 13, 14])
]),
)
assert result.shape == ()
suggested_pars = [1.1] * 3 + [0.0] * 5 # 2 pois 5 norm
constraint = poisson_constraint_combined(config)
result = default_backend.astensor(
pyhf.tensorlib.tolist(
constraint.logpdf(
pyhf.tensorlib.astensor(config.auxdata),
pyhf.tensorlib.astensor(suggested_pars),
)))
assert np.isclose(
result,
sum([
default_backend.poisson_logpdf(data, rate)
for data, rate in zip([12, 13, 14], [12 * 1.1, 13 * 1.1, 14 * 1.1])
]),
)
assert result.shape == ()
constraint = poisson_constraint_combined(config, batch_size=10)
result = constraint.logpdf(
pyhf.tensorlib.astensor(config.auxdata),
pyhf.tensorlib.astensor([suggested_pars] * 10),
)
assert result.shape == (10, )
suggested_pars = [
[1.1, 1.2, 1.3] + [0.0] * 5, # 2 pois 5 norm
[0.7, 0.8, 0.9] + [0.0] * 5, # 2 pois 5 norm
[0.4, 0.5, 0.6] + [0.0] * 5, # 2 pois 5 norm
]
constraint = poisson_constraint_combined(config, batch_size=3)
result = default_backend.astensor(
pyhf.tensorlib.tolist(
constraint.logpdf(
pyhf.tensorlib.astensor(config.auxdata),
pyhf.tensorlib.astensor(suggested_pars),
)))
assert np.all(
np.isclose(
result,
np.sum(
[
[
default_backend.poisson_logpdf(data, rate)
for data, rate in zip([12, 13, 14],
[12 * 1.1, 13 * 1.2, 14 * 1.3])
],
[
default_backend.poisson_logpdf(data, rate)
for data, rate in zip([12, 13, 14],
[12 * 0.7, 13 * 0.8, 14 * 0.9])
],
[
default_backend.poisson_logpdf(data, rate)
for data, rate in zip([12, 13, 14],
[12 * 0.4, 13 * 0.5, 14 * 0.6])
],
],
axis=1,
),
))
assert result.shape == (3, )
suggested_pars = [1.0] * 3 + [0.0] * 5 # 2 pois 5 norm
constraint = gaussian_constraint_combined(config, batch_size=1)
result = default_backend.astensor(
pyhf.tensorlib.tolist(
constraint.logpdf(
pyhf.tensorlib.astensor(config.auxdata),
pyhf.tensorlib.astensor(suggested_pars),
)))
assert np.isclose(
result[0],
sum([
default_backend.normal_logpdf(data, mu, sigma) for data, mu, sigma
in zip([0, 0, 0, 0, 0], [0, 0, 0, 0, 0], [1.5, 2.0, 1.0, 1.0, 1.0])
]),
)
assert result.shape == (1, )
suggested_pars = [1.0] * 3 + [1, 2, 3, 4, 5] # 2 pois 5 norm
constraint = gaussian_constraint_combined(config, batch_size=1)
result = default_backend.astensor(
pyhf.tensorlib.tolist(
constraint.logpdf(
pyhf.tensorlib.astensor(config.auxdata),
pyhf.tensorlib.astensor(suggested_pars),
)))
assert np.isclose(
result[0],
sum([
default_backend.normal_logpdf(data, mu, sigma) for data, mu, sigma
in zip([0, 0, 0, 0, 0], [1, 2, 3, 4, 5], [1.5, 2.0, 1.0, 1.0, 1.0])
]),
)
assert result.shape == (1, )
suggested_pars = [
[1.0] * 3 + [1, 2, 3, 4, 5], # 2 pois 5 norm
[1.0] * 3 + [-1, -2, -3, -4, -5], # 2 pois 5 norm
[1.0] * 3 + [-1, -2, 0, 1, 2], # 2 pois 5 norm
]
constraint = gaussian_constraint_combined(config, batch_size=3)
result = default_backend.astensor(
pyhf.tensorlib.tolist(
constraint.logpdf(
pyhf.tensorlib.astensor(config.auxdata),
pyhf.tensorlib.astensor(suggested_pars),
)))
assert np.all(
np.isclose(
result,
np.sum(
[
[
default_backend.normal_logpdf(data, mu, sigma)
for data, mu, sigma in
zip([0, 0, 0, 0, 0], [1, 2, 3, 4, 5],
[1.5, 2.0, 1.0, 1.0, 1.0])
],
[
default_backend.normal_logpdf(data, mu, sigma)
for data, mu, sigma in zip(
[0, 0, 0, 0, 0],
[-1, -2, -3, -4, -5],
[1.5, 2.0, 1.0, 1.0, 1.0],
)
],
[
default_backend.normal_logpdf(data, mu, sigma)
for data, mu, sigma in zip(
[0, 0, 0, 0, 0],
[-1, -2, 0, 1, 2],
[1.5, 2.0, 1.0, 1.0, 1.0],
)
],
],
axis=1,
),
))
assert result.shape == (3, )
constraint = gaussian_constraint_combined(config, batch_size=10)
result = constraint.logpdf(
pyhf.tensorlib.astensor(config.auxdata),
pyhf.tensorlib.astensor([suggested_pars] * 10),
)
assert result.shape == (10, )
def test_normsys(backend):
mc = MockConfig(
par_map={
'hello': {
'paramset': constrained_by_normal(
n_parameters=1,
inits=[0],
bounds=[[-5, 5]],
fixed=False,
auxdata=[0.0],
),
'slice': slice(0, 1),
},
'world': {
'paramset': constrained_by_normal(
n_parameters=1,
inits=[0],
bounds=[[-5, 5]],
fixed=False,
auxdata=[0.0],
),
'slice': slice(1, 2),
},
},
par_order=['hello', 'world'],
samples=['signal', 'background'],
)
mega_mods = {
'normsys/hello': {
'signal': {
'type': 'normsys',
'name': 'hello',
'data': {
'hi': [1.1] * 3,
'lo': [0.9] * 3,
'nom_data': [1, 1, 1],
'mask': [True, True, True],
},
},
'background': {
'type': 'normsys',
'name': 'hello',
'data': {
'hi': [1.2] * 3,
'lo': [0.8] * 3,
'nom_data': [1, 1, 1],
'mask': [True, True, True],
},
},
},
'normsys/world': {
'signal': {
'type': 'v',
'name': 'world',
'data': {
'hi': [1.3] * 3,
'lo': [0.7] * 3,
'nom_data': [1, 1, 1],
'mask': [True, True, True],
},
},
'background': {
'type': 'normsys',
'name': 'world',
'data': {
'hi': [1.4] * 3,
'lo': [0.6] * 3,
'nom_data': [1, 1, 1],
'mask': [True, True, True],
},
},
},
}
hsc = normsys_combined([('hello', 'normsys'), ('world', 'normsys')], mc, mega_mods)
mod = hsc.apply(pyhf.tensorlib.astensor([1.0, -1.0]))
shape = pyhf.tensorlib.shape(mod)
assert shape == (2, 2, 1, 3)
mod = np.asarray(pyhf.tensorlib.tolist(mod))
assert np.allclose(mod[0, 0, 0], [1.1, 1.1, 1.1])
assert np.allclose(mod[0, 1, 0], [1.2, 1.2, 1.2])
assert np.allclose(mod[1, 0, 0], [0.7, 0.7, 0.7])
assert np.allclose(mod[1, 1, 0], [0.6, 0.6, 0.6])
hsc = normsys_combined(
[('hello', 'normsys'), ('world', 'normsys')], mc, mega_mods, batch_size=4
)
mod = hsc.apply(
pyhf.tensorlib.astensor([[-1.0, -1.0], [1.0, 1.0], [-1.0, -1.0], [1.0, 1.0]])
)
shape = pyhf.tensorlib.shape(mod)
assert shape == (2, 2, 4, 3)
mod = np.asarray(pyhf.tensorlib.tolist(mod))
assert np.allclose(mod[0, 0, 0], [0.9, 0.9, 0.9])
assert np.allclose(mod[0, 0, 1], [1.1, 1.1, 1.1])
assert np.allclose(mod[0, 0, 2], [0.9, 0.9, 0.9])
assert np.allclose(mod[0, 0, 3], [1.1, 1.1, 1.1])
