def test_get_contour_info_nonfinite_removed(value: float) -> None:
study = prepare_study_with_trials(value_for_first_trial=value)
info = _get_contour_info(study, params=["param_b", "param_d"])
assert info == _ContourInfo(
sorted_params=["param_b", "param_d"],
sub_plot_infos=[[
_SubContourInfo(
xaxis=_AxisInfo(
name="param_b",
range=(-0.05, 1.05),
is_log=False,
is_cat=False,
indices=[-0.05, 0.0, 1.0, 1.05],
values=[0.0, 1.0],
),
yaxis=_AxisInfo(
name="param_d",
range=(1.9, 4.1),
is_log=False,
is_cat=False,
indices=[1.9, 2.0, 4.0, 4.1],
values=[4.0, 2.0],
),
z_values={
(1, 2): 2.0,
(2, 1): 1.0
},
)
]],
reverse_scale=True,
target_name="Objective Value",
)
def test_get_contour_info_2_params() -> None:
params = ["param_a", "param_b"]
study = prepare_study_with_trials()
info = _get_contour_info(study, params=params)
assert info == _ContourInfo(
sorted_params=params,
sub_plot_infos=[[
_SubContourInfo(
xaxis=_AxisInfo(
name="param_a",
range=(0.925, 2.575),
is_log=False,
is_cat=False,
indices=[0.925, 1.0, 2.5, 2.575],
values=[1.0, None, 2.5],
),
yaxis=_AxisInfo(
name="param_b",
range=(-0.1, 2.1),
is_log=False,
is_cat=False,
indices=[-0.1, 0.0, 1.0, 2.0, 2.1],
values=[2.0, 0.0, 1.0],
),
z_values={
(1, 3): 0.0,
(2, 2): 1.0
},
)
]],
reverse_scale=True,
target_name="Objective Value",
)
def test_get_contour_info_mixture_category_types() -> None:
study = _create_study_mixture_category_types()
info = _get_contour_info(study)
assert info == _ContourInfo(
sorted_params=["param_a", "param_b"],
sub_plot_infos=[[
_SubContourInfo(
xaxis=_AxisInfo(
name="param_a",
range=(-0.05, 1.05),
is_log=False,
is_cat=True,
indices=["100", "None"],
values=["None", "100"],
),
yaxis=_AxisInfo(
name="param_b",
range=(100.95, 102.05),
is_log=False,
is_cat=False,
indices=[100.95, 101, 102, 102.05],
values=[101.0, 102.0],
),
z_values={
(0, 2): 0.5,
(1, 1): 0.0
},
)
]],
reverse_scale=True,
target_name="Objective Value",
)
def test_get_contour_info_empty(specific_create_study: Callable[[], Study],
params: Optional[List[str]]) -> None:
study = specific_create_study()
info = _get_contour_info(study, params=params)
assert len(info.sorted_params) == 0
assert len(info.sub_plot_infos) == 0
def test_get_contour_info_more_than_2_params(
params: Optional[List[str]]) -> None:
study = prepare_study_with_trials()
n_params = len(params) if params is not None else 4
info = _get_contour_info(study, params=params)
assert len(info.sorted_params) == n_params
assert np.shape(np.asarray(info.sub_plot_infos)) == (n_params, n_params, 3)
def test_get_contour_info_log_scale_and_str_category_more_than_2_params(
) -> None:
# If the search space has three parameters, plot_contour generates nine plots.
study = _create_study_with_log_scale_and_str_category_3d()
info = _get_contour_info(study)
params = ["param_a", "param_b", "param_c"]
assert info.sorted_params == params
assert np.shape(np.asarray(info.sub_plot_infos)) == (3, 3, 3)
ranges = {
"param_a": (math.pow(10, -6.05), math.pow(10, -4.95)),
"param_b": (-0.05, 1.05),
"param_c": (-0.05, 1.05),
}
is_log = {"param_a": True, "param_b": False, "param_c": False}
is_cat = {"param_a": False, "param_b": True, "param_c": True}
indices: Dict[str, List[Union[str, float]]] = {
"param_a": [math.pow(10, -6.05), 1e-6, 1e-5,
math.pow(10, -4.95)],
"param_b": ["100", "101"],
"param_c": ["one", "two"],
}
values = {
"param_a": [1e-6, 1e-5],
"param_b": ["101", "100"],
"param_c": ["one", "two"]
}
def _check_axis(axis: _AxisInfo, name: str) -> None:
assert axis.name == name
assert axis.range == ranges[name]
assert axis.is_log == is_log[name]
assert axis.is_cat == is_cat[name]
assert axis.indices == indices[name]
assert axis.values == values[name]
for yi in range(3):
for xi in range(3):
xaxis = info.sub_plot_infos[yi][xi].xaxis
yaxis = info.sub_plot_infos[yi][xi].yaxis
x_param = params[xi]
y_param = params[yi]
_check_axis(xaxis, x_param)
_check_axis(yaxis, y_param)
z_values = info.sub_plot_infos[yi][xi].z_values
if xi == yi:
assert z_values == {}
else:
for i, v in enumerate([0.0, 1.0]):
x_value = xaxis.values[i]
y_value = yaxis.values[i]
assert x_value is not None
assert y_value is not None
xi = xaxis.indices.index(x_value)
yi = yaxis.indices.index(y_value)
assert z_values[(xi, yi)] == v
def test_get_contour_info_customized_target(params: List[str]) -> None:
study = prepare_study_with_trials()
info = _get_contour_info(study,
params=params,
target=lambda t: t.params["param_d"])
n_params = len(params)
assert len(info.sorted_params) == n_params
plot_shape = (1, 1, 3) if n_params == 2 else (n_params, n_params, 3)
assert np.shape(np.asarray(info.sub_plot_infos)) == plot_shape
def test_generate_contour_plot_for_few_observations(params: List[str]) -> None:
study = create_study(direction="minimize")
study.add_trial(
create_trial(
values=[0.0],
params={
"param_a": 1.0,
"param_b": 2.0
},
distributions={
"param_a": FloatDistribution(0.0, 3.0),
"param_b": FloatDistribution(0.0, 3.0),
},
))
study.add_trial(
create_trial(
values=[2.0],
params={"param_b": 0.0},
distributions={"param_b": FloatDistribution(0.0, 3.0)},
))
info = _get_contour_info(study, params=params)
assert info == _ContourInfo(
sorted_params=sorted(params),
sub_plot_infos=[[
_SubContourInfo(
xaxis=_AxisInfo(
name=sorted(params)[0],
range=(1.0, 1.0),
is_log=False,
is_cat=False,
indices=[1.0],
values=[1.0, None],
),
yaxis=_AxisInfo(
name=sorted(params)[1],
range=(-0.1, 2.1),
is_log=False,
is_cat=False,
indices=[-0.1, 0.0, 2.0, 2.1],
values=[2.0, 0.0],
),
z_values={},
)
]],
reverse_scale=True,
target_name="Objective Value",
)
def test_get_contour_info_log_scale_and_str_category_2_params() -> None:
# If the search space has two parameters, plot_contour generates a single plot.
study = _create_study_with_log_scale_and_str_category_2d()
info = _get_contour_info(study)
assert info == _ContourInfo(
sorted_params=["param_a", "param_b"],
sub_plot_infos=[[
_SubContourInfo(
xaxis=_AxisInfo(
name="param_a",
range=(math.pow(10, -6.05), math.pow(10, -4.95)),
is_log=True,
is_cat=False,
indices=[
math.pow(10, -6.05), 1e-6, 1e-5,
math.pow(10, -4.95)
],
values=[1e-6, 1e-5],
),
yaxis=_AxisInfo(
name="param_b",
range=(-0.05, 1.05),
is_log=False,
is_cat=True,
indices=["100", "101"],
values=["101", "100"],
),
z_values={
(1, 1): 0.0,
(2, 0): 1.0
},
)
]],
reverse_scale=True,
target_name="Objective Value",
)
def plot_contour(
study: Study,
params: Optional[List[str]] = None,
*,
target: Optional[Callable[[FrozenTrial], float]] = None,
target_name: str = "Objective Value",
) -> "Axes":
"""Plot the parameter relationship as contour plot in a study with Matplotlib.
Note that, if a parameter contains missing values, a trial with missing values is not plotted.
.. seealso::
Please refer to :func:`optuna.visualization.plot_contour` for an example.
Warnings:
Output figures of this Matplotlib-based
:func:`~optuna.visualization.matplotlib.plot_contour` function would be different from
those of the Plotly-based :func:`~optuna.visualization.plot_contour`.
Example:
The following code snippet shows how to plot the parameter relationship as contour plot.
.. plot::
import optuna
def objective(trial):
x = trial.suggest_float("x", -100, 100)
y = trial.suggest_categorical("y", [-1, 0, 1])
return x ** 2 + y
sampler = optuna.samplers.TPESampler(seed=10)
study = optuna.create_study(sampler=sampler)
study.optimize(objective, n_trials=30)
optuna.visualization.matplotlib.plot_contour(study, params=["x", "y"])
Args:
study:
A :class:`~optuna.study.Study` object whose trials are plotted for their target values.
params:
Parameter list to visualize. The default is all parameters.
target:
A function to specify the value to display. If it is :obj:`None` and ``study`` is being
used for single-objective optimization, the objective values are plotted.
.. note::
Specify this argument if ``study`` is being used for multi-objective optimization.
target_name:
Target's name to display on the color bar.
Returns:
A :class:`matplotlib.axes.Axes` object.
"""
_imports.check()
_logger.warning(
"Output figures of this Matplotlib-based `plot_contour` function would be different from "
"those of the Plotly-based `plot_contour`.")
info = _get_contour_info(study, params, target, target_name)
return _get_contour_plot(info)
def test_get_contour_info_too_short_params(params: List[str]) -> None:
study = prepare_study_with_trials()
info = _get_contour_info(study, params=params)
assert len(info.sorted_params) == len(params)
assert len(info.sub_plot_infos) == len(params)
def test_get_contour_info_non_exist_param_error() -> None:
study = prepare_study_with_trials()
with pytest.raises(ValueError):
_get_contour_info(study, ["optuna"])
def test_target_is_none_and_study_is_multi_obj() -> None:
study = create_study(directions=["minimize", "minimize"])
with pytest.raises(ValueError):
_get_contour_info(study)
