def manifold(self, transformer):
"""
Creates the manifold estimator if a string value is passed in,
validates other objects passed in.
"""
if not is_estimator(transformer):
if transformer not in self.ALGORITHMS:
raise YellowbrickValueError(
"could not create manifold for '%s'".format(
str(transformer)))
# Create a new transformer with the specified params
self._name = MANIFOLD_NAMES[transformer]
transformer = clone(self.ALGORITHMS[transformer])
params = {
"n_components": 2,
"n_neighbors": self.n_neighbors,
"random_state": self.random_state,
}
for param in list(params.keys()):
if param not in transformer.get_params():
del params[param]
transformer.set_params(**params)
self._manifold = transformer
if self._name is None:
self._name = self._manifold.__class__.__name__
def test_manifold_construction(self):
"""
Should be able to construct a manifold estimator from a string
"""
# TODO: parametrize this once unittest.TestCase dependency removed.
algorithms = [
"lle",
"ltsa",
"hessian",
"modified",
"isomap",
"mds",
"spectral",
"tsne",
]
for algorithm in algorithms:
message = "case failed for {}".format(algorithm)
params = {
"n_neighbors": 18,
"random_state": 53,
}
oz = Manifold(manifold=algorithm, **params)
assert is_estimator(oz.manifold), message
assert oz.manifold.get_params()["n_components"] == 2, message
manifold_params = oz.manifold.get_params()
for param, value in params.items():
if param in manifold_params:
assert value == manifold_params[param], message
def manifold(self, transformer):
"""
Creates the manifold estimator if a string value is passed in,
validates other objects passed in.
"""
if not is_estimator(transformer):
if transformer not in self.ALGORITHMS:
raise YellowbrickValueError(
"could not create manifold for '%s'".format(str(transformer))
)
# Create a new transformer with the specified params
self._name = MANIFOLD_NAMES[transformer]
transformer = clone(self.ALGORITHMS[transformer])
params = {
"n_components": 2,
"n_neighbors": self.n_neighbors,
"random_state": self.random_state,
}
for param in list(params.keys()):
if param not in transformer.get_params():
del params[param]
transformer.set_params(**params)
self._manifold = transformer
if self._name is None:
self._name = self._manifold.__class__.__name__
def get_model_name(model):
"""
Detects the model name for a Scikit-Learn model or pipeline.
Parameters
----------
model: class or instance
The object to determine the name for. If the model is an estimator it
returns the class name; if it is a Pipeline it returns the class name
of the final transformer or estimator in the Pipeline.
Returns
-------
name : string
The name of the model or pipeline.
"""
if not is_estimator(model):
raise YellowbrickTypeError(
"Cannot detect the model name for non estimator: '{}'".format(
type(model)))
if isinstance(model, Pipeline):
return get_model_name(model.steps[-1][-1])
elif isinstance(model, ContribEstimator):
return model.estimator.__class__.__name__
else:
return model.__class__.__name__
def manifold(self, transformer):
"""
Creates the manifold estimator if a string value is passed in,
validates other objects passed in.
"""
if not is_estimator(transformer):
if transformer not in self.ALGORITHMS:
raise YellowbrickValueError(
"could not create manifold for '{}'".format(
str(transformer)))
# 2 components is required for 2D plots
n_components = self.projection
requires_default_neighbors = {
"lle",
"ltsa",
"isomap",
"hessian",
"spectral",
"modified",
}
# Check if the n_neighbors attribute needs to be set.
if self.n_neighbors is None and transformer in requires_default_neighbors:
if transformer == "hessian":
self.n_neighbors = int(1 + (n_components *
(1 + (n_components + 1) / 2)))
else:
self.n_neighbors = 5
# Issue a warning that the n_neighbors was set to a default.
warnmsg = (
"using n_neighbors={};"
" please explicitly specify for the '{}' manifold").format(
self.n_neighbors, str(transformer))
warnings.warn(warnmsg, YellowbrickWarning)
# Create a new transformer with the specified params
self._name = MANIFOLD_NAMES[transformer]
transformer = clone(self.ALGORITHMS[transformer])
params = {
"n_components": n_components,
"n_neighbors": self.n_neighbors,
"random_state": self.random_state,
}
for param in list(params.keys()):
if param not in transformer.get_params():
del params[param]
transformer.set_params(**params)
self._manifold = transformer
if self._name is None:
self._name = self._manifold.__class__.__name__
def test_manifold_construction(self, algorithm):
"""
Should be able to construct a manifold estimator from a string
"""
message = "case failed for {}".format(algorithm)
params = {"n_neighbors": 18, "random_state": 53}
oz = Manifold(manifold=algorithm, **params)
assert is_estimator(oz.manifold), message
assert oz.manifold.get_params()["n_components"] == 2, message
manifold_params = oz.manifold.get_params()
for param, value in params.items():
if param in manifold_params:
assert value == manifold_params[param], message
def test_manifold_construction(self):
"""
Should be able to construct a manifold estimator from a string
"""
# TODO: parametrize this once unittest.TestCase dependency removed.
algorithms = [
"lle", "ltsa", "hessian", "modified",
"isomap", "mds", "spectral", "tsne",
]
for algorithm in algorithms:
message = "case failed for {}".format(algorithm)
params = {
"n_neighbors": 18,
"random_state": 53,
}
oz = Manifold(manifold=algorithm, **params)
assert is_estimator(oz.manifold), message
assert oz.manifold.get_params()["n_components"] == 2, message
manifold_params = oz.manifold.get_params()
for param, value in params.items():
if param in manifold_params:
assert value == manifold_params[param], message
def test_wraps_is_estimator(self):
"""
Assert a wrapped estimator passes is_estimator check
"""
tpe = wrap(ThirdPartyEstimator())
assert is_estimator(tpe)