Skip to content

update

add_core_model_to_fun_control(fun_control, core_model, hyper_dict=None, filename=None)

Add the core model to the function control dictionary. It updates the keys “core_model”, “core_model_hyper_dict”, “var_type”, “var_name” in the fun_control dictionary.

Parameters:

Name Type Description Default
fun_control dict

The fun_control dictionary.

required
core_model class

The core model.

required
hyper_dict dict

The hyper parameter dictionary. Optional. Default is None. If no hyper_dict is provided, the function will try to load the hyper_dict from the file specified by filename.

None
filename str

The name of the json file that contains the hyper parameter dictionary. Optional. Default is None. If no filename is provided, the function will try to load the hyper_dict from the hyper_dict dictionary.

None

Returns:

Type Description
dict

The updated fun_control dictionary.

Notes

The function adds the following keys to the fun_control dictionary: “core_model”: The core model. “core_model_hyper_dict”: The hyper parameter dictionary for the core model. “core_model_hyper_dict_default”: The hyper parameter dictionary for the core model. “var_type”: A list of variable types. “var_name”: A list of variable names. The original hyperparameters of the core model are stored in the “core_model_hyper_dict_default” key. These remain unmodified, while the “core_model_hyper_dict” key is modified during the tuning process.

Examples:

>>> from spotpython.light.regression.netlightregression import NetLightRegression
    from spotpython.hyperdict.light_hyper_dict import LightHyperDict
    from spotpython.hyperparameters.values import add_core_model_to_fun_control
    add_core_model_to_fun_control(fun_control=fun_control,
                                core_model=NetLightRegression,
                                hyper_dict=LightHyperDict)
    # or, if a user wants to use a custom hyper_dict:
>>> from spotpython.light.regression.netlightregression import NetLightRegression
    from spotpython.hyperparameters.values import add_core_model_to_fun_control
    add_core_model_to_fun_control(fun_control=fun_control,
                                core_model=NetLightRegression,
                                filename="./hyperdict/user_hyper_dict.json")
Source code in spotpython/hyperparameters/update.py
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
def add_core_model_to_fun_control(fun_control, core_model, hyper_dict=None, filename=None) -> dict:
    """Add the core model to the function control dictionary. It updates the keys "core_model",
    "core_model_hyper_dict", "var_type", "var_name" in the fun_control dictionary.

    Args:
        fun_control (dict):
            The fun_control dictionary.
        core_model (class):
            The core model.
        hyper_dict (dict):
            The hyper parameter dictionary. Optional. Default is None. If no hyper_dict is provided,
            the function will try to load the hyper_dict from the file specified by filename.
        filename (str):
            The name of the json file that contains the hyper parameter dictionary.
            Optional. Default is None. If no filename is provided, the function will try to load the
            hyper_dict from the hyper_dict dictionary.

    Returns:
        (dict):
            The updated fun_control dictionary.

    Notes:
        The function adds the following keys to the fun_control dictionary:
        "core_model": The core model.
        "core_model_hyper_dict": The hyper parameter dictionary for the core model.
        "core_model_hyper_dict_default": The hyper parameter dictionary for the core model.
        "var_type": A list of variable types.
        "var_name": A list of variable names.
        The original hyperparameters of the core model are stored in the "core_model_hyper_dict_default" key.
        These remain unmodified, while the "core_model_hyper_dict" key is modified during the tuning process.

    Examples:
        >>> from spotpython.light.regression.netlightregression import NetLightRegression
            from spotpython.hyperdict.light_hyper_dict import LightHyperDict
            from spotpython.hyperparameters.values import add_core_model_to_fun_control
            add_core_model_to_fun_control(fun_control=fun_control,
                                        core_model=NetLightRegression,
                                        hyper_dict=LightHyperDict)
            # or, if a user wants to use a custom hyper_dict:
        >>> from spotpython.light.regression.netlightregression import NetLightRegression
            from spotpython.hyperparameters.values import add_core_model_to_fun_control
            add_core_model_to_fun_control(fun_control=fun_control,
                                        core_model=NetLightRegression,
                                        filename="./hyperdict/user_hyper_dict.json")

    """
    fun_control.update({"core_model": core_model})
    if filename is None:
        new_hyper_dict = hyper_dict().load()
    else:
        with open(filename, "r") as f:
            new_hyper_dict = json.load(f)
    fun_control.update({"core_model_hyper_dict": new_hyper_dict[core_model.__name__]})
    fun_control.update({"core_model_hyper_dict_default": copy.deepcopy(new_hyper_dict[core_model.__name__])})
    var_type = get_var_type(fun_control)
    var_name = get_var_name(fun_control)
    lower = get_bound_values(fun_control, "lower", as_list=False)
    upper = get_bound_values(fun_control, "upper", as_list=False)
    fun_control.update({"var_type": var_type, "var_name": var_name, "lower": lower, "upper": upper})

assign_values(X, var_list)

This function takes an np.array X and a list of variable names as input arguments and returns a dictionary where the keys are the variable names and the values are assigned from X.

Parameters:

Name Type Description Default
X array

A 2D numpy array where each column represents a variable.

required
var_list list

A list of strings representing variable names.

required

Returns:

Name Type Description
dict dict

A dictionary where keys are variable names and values are assigned from X.

Examples:

>>> import numpy as np
>>> from spotpython.hyperparameters.values import assign_values
>>> X = np.array([[1, 2], [3, 4], [5, 6]])
>>> var_list = ['a', 'b']
>>> result = assign_values(X, var_list)
>>> print(result)
{'a': array([1, 3, 5]), 'b': array([2, 4, 6])}
Source code in spotpython/hyperparameters/update.py
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
def assign_values(X: np.array, var_list: list) -> dict:
    """
    This function takes an np.array X and a list of variable names as input arguments
    and returns a dictionary where the keys are the variable names and the values are assigned from X.

    Args:
        X (np.array):
            A 2D numpy array where each column represents a variable.
        var_list (list):
            A list of strings representing variable names.

    Returns:
        dict:
            A dictionary where keys are variable names and values are assigned from X.

    Examples:
        >>> import numpy as np
        >>> from spotpython.hyperparameters.values import assign_values
        >>> X = np.array([[1, 2], [3, 4], [5, 6]])
        >>> var_list = ['a', 'b']
        >>> result = assign_values(X, var_list)
        >>> print(result)
        {'a': array([1, 3, 5]), 'b': array([2, 4, 6])}
    """
    result = {}
    for i, var_name in enumerate(var_list):
        result[var_name] = X[:, i]
    return result

convert_keys(d, var_type)

Convert values in a dictionary to integers based on a list of variable types. This function takes a dictionary d and a list of variable types var_type as arguments. For each key in the dictionary, if the corresponding entry in var_type is not equal to "num", the value associated with that key is converted to an integer.

Parameters:

Name Type Description Default
d dict

The input dictionary.

required
var_type list

A list of variable types. If the entry is not "num" the corresponding value will be converted to the type "int".

required

Returns:

Name Type Description
dict Dict[str, Union[int, float]]

The modified dictionary with values converted to integers based on var_type.

Examples:

>>> from spotpython.hyperparameters.values import convert_keys
>>> d = {'a': '1.1', 'b': '2', 'c': '3.1'}
>>> var_type = ["int", "num", "int"]
>>> convert_keys(d, var_type)
{'a': 1, 'b': '2', 'c': 3}
Source code in spotpython/hyperparameters/update.py
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
def convert_keys(d: Dict[str, Union[int, float, str]], var_type: List[str]) -> Dict[str, Union[int, float]]:
    """Convert values in a dictionary to integers based on a list of variable types.
    This function takes a dictionary `d` and a list of variable types `var_type` as arguments.
    For each key in the dictionary,
    if the corresponding entry in `var_type` is not equal to `"num"`,
    the value associated with that key is converted to an integer.

    Args:
        d (dict): The input dictionary.
        var_type (list):
            A list of variable types. If the entry is not `"num"` the corresponding
            value will be converted to the type `"int"`.

    Returns:
        dict: The modified dictionary with values converted to integers based on `var_type`.

    Examples:
        >>> from spotpython.hyperparameters.values import convert_keys
        >>> d = {'a': '1.1', 'b': '2', 'c': '3.1'}
        >>> var_type = ["int", "num", "int"]
        >>> convert_keys(d, var_type)
        {'a': 1, 'b': '2', 'c': 3}
    """
    keys = list(d.keys())
    for i in range(len(keys)):
        if var_type[i] not in ["num", "float"]:
            d[keys[i]] = int(d[keys[i]])
    return d

create_model(config, fun_control, **kwargs)

Creates a model for the given configuration and control parameters.

Parameters:

Name Type Description Default
config dict

dictionary containing the configuration for the hyperparameter tuning.

required
fun_control dict

dictionary containing control parameters for the hyperparameter tuning.

required
**kwargs Any

additional keyword arguments.

{}

Returns:

Type Description
object

model object.

Source code in spotpython/hyperparameters/update.py
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
def create_model(config, fun_control, **kwargs) -> object:
    """
    Creates a model for the given configuration and control parameters.

    Args:
        config (dict):
            dictionary containing the configuration for the hyperparameter tuning.
        fun_control (dict):
            dictionary containing control parameters for the hyperparameter tuning.
        **kwargs (Any):
            additional keyword arguments.

    Returns:
        (object):
            model object.
    """
    return fun_control["core_model"](**config, **kwargs)

generate_one_config_from_var_dict(var_dict, fun_control, default=False)

Generate one configuration from a dictionary of variables (as a generator).

This function takes a dictionary of variables as input arguments and returns a generator that yields dictionaries with the values from the arrays in the input dictionary.

Parameters:

Name Type Description Default
var_dict dict

A dictionary where keys are variable names and values are numpy arrays.

required
fun_control dict

A dictionary which (at least) has an entry with the following key: “var_type” (list): A list of variable types. If the entry is not “num” the corresponding value will be converted to the type “int”.

required
default bool

A boolean value indicating whether to use the default values from fun_control.

False

Returns:

Type Description
Generator[Dict[str, Union[int, float]], None, None]

Generator[dict]: A generator that yields dictionaries with the values from the arrays in the input dictionary.

Examples:

>>> import numpy as np
>>> from spotpython.hyperparameters.values import generate_one_config_from_var_dict
>>> var_dict = {'a': np.array([1, 3, 5]), 'b': np.array([2, 4, 6])}
>>> fun_control = {"var_type": ["int", "num"]}
>>> list(generate_one_config_from_var_dict(var_dict, fun_control))
[{'a': 1, 'b': 2}, {'a': 3, 'b': 4}, {'a': 5, 'b': 6}]
Source code in spotpython/hyperparameters/update.py
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
def generate_one_config_from_var_dict(
    var_dict: Dict[str, np.ndarray],
    fun_control: Dict[str, Union[List[str], str]],
    default: bool = False,
) -> Generator[Dict[str, Union[int, float]], None, None]:
    """Generate one configuration from a dictionary of variables (as a generator).

    This function takes a dictionary of variables as input arguments and returns a generator
    that yields dictionaries with the values from the arrays in the input dictionary.

    Args:
        var_dict (dict):
            A dictionary where keys are variable names and values are numpy arrays.
        fun_control (dict):
            A dictionary which (at least) has an entry with the following key:
            "var_type" (list): A list of variable types. If the entry is not "num" the corresponding
            value will be converted to the type "int".
        default (bool):
            A boolean value indicating whether to use the default values from fun_control.

    Returns:
        Generator[dict]: A generator that yields dictionaries with the values from the arrays in the input dictionary.

    Examples:
        >>> import numpy as np
        >>> from spotpython.hyperparameters.values import generate_one_config_from_var_dict
        >>> var_dict = {'a': np.array([1, 3, 5]), 'b': np.array([2, 4, 6])}
        >>> fun_control = {"var_type": ["int", "num"]}
        >>> list(generate_one_config_from_var_dict(var_dict, fun_control))
        [{'a': 1, 'b': 2}, {'a': 3, 'b': 4}, {'a': 5, 'b': 6}]
    """
    for values in iterate_dict_values(var_dict):
        values = convert_keys(values, fun_control["var_type"])
        values = get_dict_with_levels_and_types(fun_control=fun_control, v=values, default=default)
        values = transform_hyper_parameter_values(fun_control=fun_control, hyper_parameter_values=values)
        yield values

get_bound_values(fun_control, bound, as_list=False)

Generate a list or array from a dictionary. This function takes the values from the keys “bound” in the fun_control[“core_model_hyper_dict”] dictionary and returns a list or array of the values in the same order as the keys in the dictionary.

Parameters:

Name Type Description Default
fun_control dict

A dictionary containing a key “core_model_hyper_dict” which is a dictionary with keys that have either an “upper” or “lower” value.

required
bound str

Either “upper” or “lower”, indicating which value to extract from the inner dictionary.

required
as_list bool

If True, return a list. If False, return a numpy array. Default is False.

False

Returns:

Type Description
Union[List, ndarray]

list or np.ndarray: A list or array of the extracted values.

Raises:

Type Description
ValueError

If bound is not “upper” or “lower”.

Examples:

>>> from spotpython.hyperparameters.values import get_bound_values
>>> fun_control = {"core_model_hyper_dict": {"a": {"upper": 1}, "b": {"upper": 2}}}
>>> get_bound_values(fun_control, "upper", as_list=True)
[1, 2]
Source code in spotpython/hyperparameters/update.py
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
def get_bound_values(fun_control: dict, bound: str, as_list: bool = False) -> Union[List, np.ndarray]:
    """Generate a list or array from a dictionary.
    This function takes the values from the keys "bound" in the
    fun_control["core_model_hyper_dict"] dictionary and returns a list or array of the values
    in the same order as the keys in the dictionary.

    Args:
        fun_control (dict):
            A dictionary containing a key "core_model_hyper_dict"
            which is a dictionary with keys that have either an "upper" or "lower" value.
        bound (str):
            Either "upper" or "lower",
            indicating which value to extract from the inner dictionary.
        as_list (bool):
            If True, return a list.
            If False, return a numpy array. Default is False.

    Returns:
        list or np.ndarray:
            A list or array of the extracted values.

    Raises:
        ValueError:
            If bound is not "upper" or "lower".

    Examples:
        >>> from spotpython.hyperparameters.values import get_bound_values
        >>> fun_control = {"core_model_hyper_dict": {"a": {"upper": 1}, "b": {"upper": 2}}}
        >>> get_bound_values(fun_control, "upper", as_list=True)
        [1, 2]
    """
    # Throw value error if bound is not upper or lower:
    if bound not in ["upper", "lower"]:
        raise ValueError("bound must be either 'upper' or 'lower'")
    # check if key "core_model_hyper_dict" exists in fun_control:
    if "core_model_hyper_dict" not in fun_control.keys():
        return None
    else:
        d = fun_control["core_model_hyper_dict"]
        b = []
        for key, value in d.items():
            b.append(value[bound])
        if as_list:
            return b
        else:
            return np.array(b)

get_control_key_value(control_dict=None, key=None)

This function gets the key value pair from the control_dict dictionary. If the key does not exist, return None. If the control_dict dictionary is None, return None.

Parameters:

Name Type Description Default
control_dict dict

control_dict dictionary

None
key str

key

None

Returns:

Name Type Description
value Any

value

Examples:

>>> from spotpython.utils.init import fun_control_init
    from spotpython.hyperparameters.values import get_control_key_value
    control_dict = fun_control_init()
    get_control_key_value(control_dict=control_dict,
                    key="key")
    "value"
Source code in spotpython/hyperparameters/update.py
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
def get_control_key_value(control_dict=None, key=None) -> Any:
    """
    This function gets the key value pair from the control_dict dictionary.
    If the key does not exist, return None.
    If the control_dict dictionary is None, return None.

    Args:
        control_dict (dict):
            control_dict dictionary
        key (str): key

    Returns:
        value (Any):
            value

    Examples:
        >>> from spotpython.utils.init import fun_control_init
            from spotpython.hyperparameters.values import get_control_key_value
            control_dict = fun_control_init()
            get_control_key_value(control_dict=control_dict,
                            key="key")
            "value"
    """
    if control_dict is None:
        return None
    else:
        # check if key "core_model_hyper_dict" exists in fun_control:
        if "core_model_hyper_dict" in control_dict.keys():
            if key == "lower":
                lower = get_bound_values(fun_control=control_dict, bound="lower")
                return lower
            if key == "upper":
                upper = get_bound_values(fun_control=control_dict, bound="upper")
                return upper
            if key == "var_name":
                var_name = get_var_name(fun_control=control_dict)
                return var_name
            if key == "var_type":
                var_type = get_var_type(fun_control=control_dict)
                return var_type
            if key == "transform":
                transform = get_transform(fun_control=control_dict)
                return transform
        # check if key exists in control_dict:
        elif control_dict is None or key not in control_dict.keys():
            return None
        else:
            return control_dict[key]

get_core_model_parameter_type_from_var_name(fun_control, var_name)

Extracts the core_model_parameter_type value from a dictionary for a specified key.

Parameters:

Name Type Description Default
fun_control dict

The dictionary containing the information.

required
var_name str

The key for which to extract the core_model_parameter_type value.

required

Returns:

Type Description
str

The core_model_parameter_type value if available, else None.

Source code in spotpython/hyperparameters/update.py
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
def get_core_model_parameter_type_from_var_name(fun_control, var_name) -> str:
    """
    Extracts the core_model_parameter_type value from a dictionary for a specified key.

    Args:
        fun_control (dict):
            The dictionary containing the information.
        var_name (str):
            The key for which to extract the core_model_parameter_type value.

    Returns:
        (str):
            The core_model_parameter_type value if available, else None.
    """
    # Check if the key exists in the dictionary and it has a 'core_model_parameter_type' entry
    if var_name in fun_control["core_model_hyper_dict"] and "core_model_parameter_type" in fun_control["core_model_hyper_dict"][var_name]:
        return fun_control["core_model_hyper_dict"][var_name]["core_model_parameter_type"]
    else:
        return None

get_default_hyperparameters_as_array(fun_control)

Get the default hyper parameters as array.

Parameters:

Name Type Description Default
fun_control dict

The function control dictionary.

required

Returns:

Type Description
array

The default hyper parameters as array.

Examples:

>>> from river.tree import HoeffdingAdaptiveTreeRegressor
    from spotriver.data.river_hyper_dict import RiverHyperDict
    from spotpython.hyperparameters.values import (
        get_default_hyperparameters_as_array,
        add_core_model_to_fun_control)
    fun_control = {}
    add_core_model_to_fun_control(core_model=HoeffdingAdaptiveTreeRegressor,
        fun_control=func_control,
        hyper_dict=RiverHyperDict,
        filename=None)
    get_default_hyperparameters_as_array(fun_control)
    array([0, 0, 0, 0, 0])
Source code in spotpython/hyperparameters/update.py
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
def get_default_hyperparameters_as_array(fun_control) -> np.array:
    """Get the default hyper parameters as array.

    Args:
        fun_control (dict):
            The function control dictionary.

    Returns:
        (np.array):
            The default hyper parameters as array.

    Examples:
        >>> from river.tree import HoeffdingAdaptiveTreeRegressor
            from spotriver.data.river_hyper_dict import RiverHyperDict
            from spotpython.hyperparameters.values import (
                get_default_hyperparameters_as_array,
                add_core_model_to_fun_control)
            fun_control = {}
            add_core_model_to_fun_control(core_model=HoeffdingAdaptiveTreeRegressor,
                fun_control=func_control,
                hyper_dict=RiverHyperDict,
                filename=None)
            get_default_hyperparameters_as_array(fun_control)
            array([0, 0, 0, 0, 0])
    """
    X0 = get_default_values(fun_control)
    X0 = replace_levels_with_positions(fun_control["core_model_hyper_dict_default"], X0)
    if X0 is None:
        return None
    else:
        X0 = get_values_from_dict(X0)
        X0 = np.array([X0])
        X0.shape[1]
        return X0

get_default_values(fun_control)

Get the values from the “default” keys from the dictionary fun_control as a dict. If the key of the value has as “type” the value “int” or “float”, convert the value to the corresponding type.

Parameters:

Name Type Description Default
fun_control dict

dictionary with levels and types

required

Returns:

Name Type Description
new_dict dict

dictionary with default values

Examples:

>>> from spotpython.hyperparameters.values import get_default_values
    d = {"core_model_hyper_dict":{
        "leaf_prediction": {
            "levels": ["mean", "model", "adaptive"],
            "type": "factor",
            "default": "mean",
            "core_model_parameter_type": "str"},
        "leaf_model": {
            "levels": ["linear_model.LinearRegression", "linear_model.PARegressor", "linear_model.Perceptron"],
            "type": "factor",
            "default": "LinearRegression",
            "core_model_parameter_type": "instance"},
        "splitter": {
            "levels": ["EBSTSplitter", "TEBSTSplitter", "QOSplitter"],
            "type": "factor",
            "default": "EBSTSplitter",
            "core_model_parameter_type": "instance()"},
        "binary_split": {
            "levels": [0, 1],
            "type": "factor",
            "default": 0,
            "core_model_parameter_type": "bool"},
        "stop_mem_management": {
            "levels": [0, 1],
            "type": "factor",
            "default": 0,
            "core_model_parameter_type": "bool"}}}
    get_default_values(d)
    {'leaf_prediction': 'mean',
    'leaf_model': 'linear_model.LinearRegression',
    'splitter': 'EBSTSplitter',
    'binary_split': 0,
    'stop_mem_management': 0}
Source code in spotpython/hyperparameters/update.py
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
def get_default_values(fun_control) -> dict:
    """Get the values from the "default" keys from the dictionary fun_control as a dict.
    If the key of the value has as "type" the value "int" or "float", convert the value to the corresponding type.

    Args:
        fun_control (dict):
            dictionary with levels and types

    Returns:
        new_dict (dict):
            dictionary with default values

    Examples:
        >>> from spotpython.hyperparameters.values import get_default_values
            d = {"core_model_hyper_dict":{
                "leaf_prediction": {
                    "levels": ["mean", "model", "adaptive"],
                    "type": "factor",
                    "default": "mean",
                    "core_model_parameter_type": "str"},
                "leaf_model": {
                    "levels": ["linear_model.LinearRegression", "linear_model.PARegressor", "linear_model.Perceptron"],
                    "type": "factor",
                    "default": "LinearRegression",
                    "core_model_parameter_type": "instance"},
                "splitter": {
                    "levels": ["EBSTSplitter", "TEBSTSplitter", "QOSplitter"],
                    "type": "factor",
                    "default": "EBSTSplitter",
                    "core_model_parameter_type": "instance()"},
                "binary_split": {
                    "levels": [0, 1],
                    "type": "factor",
                    "default": 0,
                    "core_model_parameter_type": "bool"},
                "stop_mem_management": {
                    "levels": [0, 1],
                    "type": "factor",
                    "default": 0,
                    "core_model_parameter_type": "bool"}}}
            get_default_values(d)
            {'leaf_prediction': 'mean',
            'leaf_model': 'linear_model.LinearRegression',
            'splitter': 'EBSTSplitter',
            'binary_split': 0,
            'stop_mem_management': 0}
    """
    d = fun_control["core_model_hyper_dict_default"]
    new_dict = {}
    for key, value in d.items():
        if value["type"] == "int":
            new_dict[key] = int(value["default"])
        elif value["type"] == "float":
            new_dict[key] = float(value["default"])
        else:
            new_dict[key] = value["default"]
    return new_dict

get_dict_with_levels_and_types(fun_control, v, default=False)

Get dictionary with levels and types. The function maps the numerical output of the hyperparameter optimization to the corresponding levels of the hyperparameter needed by the core model, i.e., the tuned algorithm. The function takes the dictionaries fun_control and v and returns a new dictionary with the same keys as v but with the values of the levels of the keys from fun_control. If the key value in the dictionary is 0, it takes the first value from the list, if it is 1, it takes the second and so on. If a key is not in fun_control, it takes the key from v. If the core_model_parameter_type value is instance, it returns the class of the value from the module via getattr(“class”, value).

Parameters:

Name Type Description Default
fun_control Dict[str, Any]

A dictionary containing information about the core model hyperparameters.

required
v Dict[str, Any]

A dictionary containing the numerical output of the hyperparameter optimization.

required
default bool

A boolean value indicating whether to use the default values from fun_control.

False

Returns:

Type Description
Dict[str, Any]

Dict[str, Any]: A new dictionary with the same keys as v but with the values of the levels of the keys from fun_control.

Examples:

>>> fun_control = {
...     "core_model_hyper_dict": {
...         "leaf_prediction": {
...             "levels": ["mean", "model", "adaptive"],
...             "type": "factor",
...             "default": "mean",
...             "core_model_parameter_type": "str"
...         },
...         "leaf_model": {
...             "levels": [
...                 "linear_model.LinearRegression",
...                 "linear_model.PARegressor",
...                 "linear_model.Perceptron"
...             ],
...             "type": "factor",
...             "default": "LinearRegression",
...             "core_model_parameter_type": "instance"
...         },
...         "splitter": {
...             "levels": ["EBSTSplitter", "TEBSTSplitter", "QOSplitter"],
...             "type": "factor",
...             "default": "EBSTSplitter",
...             "core_model_parameter_type": "instance()"
...         },
...         "binary_split": {
...             "levels": [0, 1],
...             "type": "factor",
...             "default": 0,
...             "core_model_parameter_type": "bool"
...         },
...         "stop_mem_management": {
...             "levels": [0, 1],
...             "type": "factor",
...             "default": 0,
...             "core_model_parameter_type": "bool"
...         }
...     }
... }
>>> v = {
...     'grace_period': 200,
...     'max_depth': 10,
...     'delta': 1e-07,
...     'tau': 0.05,
...     'leaf_prediction': 0,
...     'leaf_model': 0,
...     'model_selector_decay': 0.95,
...     'splitter': 1,
...     'min_samples_split': 9,
...     'binary_split': 0,
...     'max_size': 500.0
... }
>>> get_dict_with_levels_and_types(fun_control, v)
{
    'grace_period': 200,
    'max_depth': 10,
    'delta': 1e-07,
    'tau': 0.05,
    'leaf_prediction': 'mean',
    'leaf_model': linear_model.LinearRegression,
    'model_selector_decay': 0.95,
    'splitter': TEBSTSplitter,
    'min_samples_split': 9,
    'binary_split': False,
    'max_size': 500.0
}
Source code in spotpython/hyperparameters/update.py
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
def get_dict_with_levels_and_types(fun_control: Dict[str, Any], v: Dict[str, Any], default=False) -> Dict[str, Any]:
    """Get dictionary with levels and types.
    The function maps the numerical output of the hyperparameter optimization to the corresponding levels
    of the hyperparameter needed by the core model, i.e., the tuned algorithm.
    The function takes the dictionaries fun_control and v and returns a new dictionary with the same keys as v
    but with the values of the levels of the keys from fun_control.
    If the key value in the dictionary is 0, it takes the first value from the list,
    if it is 1, it takes the second and so on.
    If a key is not in fun_control, it takes the key from v.
    If the core_model_parameter_type value is instance, it returns the class of the value from the module
    via getattr("class", value).

    Args:
        fun_control (Dict[str, Any]):
            A dictionary containing information about the core model hyperparameters.
        v (Dict[str, Any]):
            A dictionary containing the numerical output of the hyperparameter optimization.
        default (bool):
            A boolean value indicating whether to use the default values from fun_control.

    Returns:
        Dict[str, Any]:
            A new dictionary with the same keys as v but with the values of the levels of the keys from fun_control.

    Examples:
        >>> fun_control = {
        ...     "core_model_hyper_dict": {
        ...         "leaf_prediction": {
        ...             "levels": ["mean", "model", "adaptive"],
        ...             "type": "factor",
        ...             "default": "mean",
        ...             "core_model_parameter_type": "str"
        ...         },
        ...         "leaf_model": {
        ...             "levels": [
        ...                 "linear_model.LinearRegression",
        ...                 "linear_model.PARegressor",
        ...                 "linear_model.Perceptron"
        ...             ],
        ...             "type": "factor",
        ...             "default": "LinearRegression",
        ...             "core_model_parameter_type": "instance"
        ...         },
        ...         "splitter": {
        ...             "levels": ["EBSTSplitter", "TEBSTSplitter", "QOSplitter"],
        ...             "type": "factor",
        ...             "default": "EBSTSplitter",
        ...             "core_model_parameter_type": "instance()"
        ...         },
        ...         "binary_split": {
        ...             "levels": [0, 1],
        ...             "type": "factor",
        ...             "default": 0,
        ...             "core_model_parameter_type": "bool"
        ...         },
        ...         "stop_mem_management": {
        ...             "levels": [0, 1],
        ...             "type": "factor",
        ...             "default": 0,
        ...             "core_model_parameter_type": "bool"
        ...         }
        ...     }
        ... }
        >>> v = {
        ...     'grace_period': 200,
        ...     'max_depth': 10,
        ...     'delta': 1e-07,
        ...     'tau': 0.05,
        ...     'leaf_prediction': 0,
        ...     'leaf_model': 0,
        ...     'model_selector_decay': 0.95,
        ...     'splitter': 1,
        ...     'min_samples_split': 9,
        ...     'binary_split': 0,
        ...     'max_size': 500.0
        ... }
        >>> get_dict_with_levels_and_types(fun_control, v)
        {
            'grace_period': 200,
            'max_depth': 10,
            'delta': 1e-07,
            'tau': 0.05,
            'leaf_prediction': 'mean',
            'leaf_model': linear_model.LinearRegression,
            'model_selector_decay': 0.95,
            'splitter': TEBSTSplitter,
            'min_samples_split': 9,
            'binary_split': False,
            'max_size': 500.0
        }
    """
    if default:
        d = fun_control["core_model_hyper_dict_default"]
    else:
        d = fun_control["core_model_hyper_dict"]
    new_dict = {}
    for key, value in v.items():
        if key in d and d[key]["type"] == "factor":
            if d[key]["core_model_parameter_type"] == "instance":
                if "class_name" in d[key]:
                    mdl = d[key]["class_name"]
                c = d[key]["levels"][value]
                new_dict[key] = class_for_name(mdl, c)
            elif d[key]["core_model_parameter_type"] == "instance()":
                mdl = d[key]["class_name"]
                c = d[key]["levels"][value]
                k = class_for_name(mdl, c)
                new_dict[key] = k()
            else:
                new_dict[key] = d[key]["levels"][value]
        else:
            new_dict[key] = v[key]
    return new_dict

get_ith_hyperparameter_name_from_fun_control(fun_control, key, i)

Get the ith hyperparameter name from the fun_control dictionary.

Parameters:

Name Type Description Default
fun_control dict

fun_control dictionary

required
key str

key

required
i int

index

required

Returns:

Type Description
str

hyperparameter name

Examples:

>>> from spotpython.utils.device import getDevice
    from spotpython.utils.init import fun_control_init
    from spotpython.utils.file import get_experiment_name
    import numpy as np
    from spotpython.data.diabetes import Diabetes
    from spotpython.light.regression.netlightregression import NetLightRegression
    from spotpython.hyperdict.light_hyper_dict import LightHyperDict
    from spotpython.hyperparameters.values import add_core_model_to_fun_control
    from spotpython.hyperparameters.values import get_ith_hyperparameter_name_from_fun_control
    from spotpython.hyperparameters.values import set_control_key_value
    from spotpython.hyperparameters.values import set_control_hyperparameter_value
    experiment_name = get_experiment_name(prefix="000")
    fun_control = fun_control_init(
        _L_in=10,
        _L_out=1,
        TENSORBOARD_CLEAN=True,
        device=getDevice(),
        enable_progress_bar=False,
        fun_evals=15,
        log_level=10,
        max_time=1,
        num_workers=0,
        show_progress=True,
        tolerance_x=np.sqrt(np.spacing(1)),
        )
    dataset = Diabetes()
    set_control_key_value(control_dict=fun_control,
                            key="data_set",
                            value=dataset,
                            replace=True)
    add_core_model_to_fun_control(core_model=NetLightRegression,
                                fun_control=fun_control,
                                hyper_dict=LightHyperDict)
set_control_hyperparameter_value(fun_control, "l1", [3,8])
set_control_hyperparameter_value(fun_control, "optimizer", ["Adam", "AdamW", "Adamax", "NAdam"])
get_ith_hyperparameter_name_from_fun_control(fun_control, key="optimizer", i=0)
Adam
Source code in spotpython/hyperparameters/update.py
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
def get_ith_hyperparameter_name_from_fun_control(fun_control, key, i):
    """
    Get the ith hyperparameter name from the fun_control dictionary.

    Args:
        fun_control (dict): fun_control dictionary
        key (str): key
        i (int): index

    Returns:
        (str): hyperparameter name

    Examples:
        >>> from spotpython.utils.device import getDevice
            from spotpython.utils.init import fun_control_init
            from spotpython.utils.file import get_experiment_name
            import numpy as np
            from spotpython.data.diabetes import Diabetes
            from spotpython.light.regression.netlightregression import NetLightRegression
            from spotpython.hyperdict.light_hyper_dict import LightHyperDict
            from spotpython.hyperparameters.values import add_core_model_to_fun_control
            from spotpython.hyperparameters.values import get_ith_hyperparameter_name_from_fun_control
            from spotpython.hyperparameters.values import set_control_key_value
            from spotpython.hyperparameters.values import set_control_hyperparameter_value
            experiment_name = get_experiment_name(prefix="000")
            fun_control = fun_control_init(
                _L_in=10,
                _L_out=1,
                TENSORBOARD_CLEAN=True,
                device=getDevice(),
                enable_progress_bar=False,
                fun_evals=15,
                log_level=10,
                max_time=1,
                num_workers=0,
                show_progress=True,
                tolerance_x=np.sqrt(np.spacing(1)),
                )
            dataset = Diabetes()
            set_control_key_value(control_dict=fun_control,
                                    key="data_set",
                                    value=dataset,
                                    replace=True)
            add_core_model_to_fun_control(core_model=NetLightRegression,
                                        fun_control=fun_control,
                                        hyper_dict=LightHyperDict)

            set_control_hyperparameter_value(fun_control, "l1", [3,8])
            set_control_hyperparameter_value(fun_control, "optimizer", ["Adam", "AdamW", "Adamax", "NAdam"])
            get_ith_hyperparameter_name_from_fun_control(fun_control, key="optimizer", i=0)
            Adam

    """
    if "core_model_hyper_dict" in fun_control:
        if key in fun_control["core_model_hyper_dict"]:
            if "levels" in fun_control["core_model_hyper_dict"][key]:
                if i < len(fun_control["core_model_hyper_dict"][key]["levels"]):
                    return fun_control["core_model_hyper_dict"][key]["levels"][i]
    return None

get_one_config_from_X(X, fun_control=None)

Get one config from X.

Parameters:

Name Type Description Default
X array

The array with the hyper parameter values.

required
fun_control dict

The function control dictionary.

None

Returns:

Type Description
dict

The config dictionary.

Examples:

>>> from river.tree import HoeffdingAdaptiveTreeRegressor
    from spotriver.data.river_hyper_dict import RiverHyperDict
    fun_control = {}
    add_core_model_to_fun_control(core_model=HoeffdingAdaptiveTreeRegressor,
        fun_control=func_control,
        hyper_dict=RiverHyperDict,
        filename=None)
    X = np.array([0, 0, 0, 0, 0])
    get_one_config_from_X(X, fun_control)
    {'leaf_prediction': 'mean',
    'leaf_model': 'NBAdaptive',
    'splitter': 'HoeffdingAdaptiveTreeSplitter',
    'binary_split': 'info_gain',
    'stop_mem_management': False}
Source code in spotpython/hyperparameters/update.py
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
def get_one_config_from_X(X, fun_control=None):
    """Get one config from X.

    Args:
        X (np.array):
            The array with the hyper parameter values.
        fun_control (dict):
            The function control dictionary.

    Returns:
        (dict):
            The config dictionary.

    Examples:
        >>> from river.tree import HoeffdingAdaptiveTreeRegressor
            from spotriver.data.river_hyper_dict import RiverHyperDict
            fun_control = {}
            add_core_model_to_fun_control(core_model=HoeffdingAdaptiveTreeRegressor,
                fun_control=func_control,
                hyper_dict=RiverHyperDict,
                filename=None)
            X = np.array([0, 0, 0, 0, 0])
            get_one_config_from_X(X, fun_control)
            {'leaf_prediction': 'mean',
            'leaf_model': 'NBAdaptive',
            'splitter': 'HoeffdingAdaptiveTreeSplitter',
            'binary_split': 'info_gain',
            'stop_mem_management': False}
    """
    var_dict = assign_values(X, fun_control["var_name"])
    config = return_conf_list_from_var_dict(var_dict, fun_control)[0]
    return config

get_one_core_model_from_X(X, fun_control=None, default=False)

Get one core model from X.

Parameters:

Name Type Description Default
X array

The array with the hyper parameter values.

required
fun_control dict

The function control dictionary.

None
default bool

A boolean value indicating whether to use the default values from fun_control.

False

Returns:

Type Description
class

The core model.

Examples:

>>> from river.tree import HoeffdingAdaptiveTreeRegressor
    from spotriver.data.river_hyper_dict import RiverHyperDict
    fun_control = {}
    add_core_model_to_fun_control(core_model=HoeffdingAdaptiveTreeRegressor,
        fun_control=fun_control,
        hyper_dict=RiverHyperDict,
        filename=None)
    X = np.array([0, 0, 0, 0, 0])
    get_one_core_model_from_X(X, fun_control)
    HoeffdingAdaptiveTreeRegressor()
Source code in spotpython/hyperparameters/update.py
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
def get_one_core_model_from_X(
    X,
    fun_control=None,
    default=False,
):
    """Get one core model from X.

    Args:
        X (np.array):
            The array with the hyper parameter values.
        fun_control (dict):
            The function control dictionary.
        default (bool):
            A boolean value indicating whether to use the default values from fun_control.

    Returns:
        (class):
            The core model.

    Examples:
        >>> from river.tree import HoeffdingAdaptiveTreeRegressor
            from spotriver.data.river_hyper_dict import RiverHyperDict
            fun_control = {}
            add_core_model_to_fun_control(core_model=HoeffdingAdaptiveTreeRegressor,
                fun_control=fun_control,
                hyper_dict=RiverHyperDict,
                filename=None)
            X = np.array([0, 0, 0, 0, 0])
            get_one_core_model_from_X(X, fun_control)
            HoeffdingAdaptiveTreeRegressor()
    """
    var_dict = assign_values(X, fun_control["var_name"])
    # var_dict = assign_values(X, get_var_name(fun_control))
    config = return_conf_list_from_var_dict(var_dict, fun_control, default=default)[0]
    core_model = fun_control["core_model"](**config)
    return core_model

get_one_river_model_from_X(X, fun_control=None)

Get one river model from X.

Parameters:

Name Type Description Default
X array

The array with the hyper parameter values.

required
fun_control dict

The function control dictionary.

None

Returns:

Type Description
class

The river model.

Examples:

>>> from river.tree import HoeffdingAdaptiveTreeRegressor
    from spotriver.data.river_hyper_dict import RiverHyperDict
    fun_control = {}
    add_core_model_to_fun_control(core_model=HoeffdingAdaptiveTreeRegressor,
        fun_control=func_control,
        hyper_dict=RiverHyperDict,
        filename=None)
    X = np.array([0, 0, 0, 0, 0])
    get_one_river_model_from_X(X, fun_control)
    HoeffdingAdaptiveTreeRegressor()
Source code in spotpython/hyperparameters/update.py
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
def get_one_river_model_from_X(X, fun_control=None):
    """Get one river model from X.

    Args:
        X (np.array):
            The array with the hyper parameter values.
        fun_control (dict):
            The function control dictionary.

    Returns:
        (class):
            The river model.

    Examples:
        >>> from river.tree import HoeffdingAdaptiveTreeRegressor
            from spotriver.data.river_hyper_dict import RiverHyperDict
            fun_control = {}
            add_core_model_to_fun_control(core_model=HoeffdingAdaptiveTreeRegressor,
                fun_control=func_control,
                hyper_dict=RiverHyperDict,
                filename=None)
            X = np.array([0, 0, 0, 0, 0])
            get_one_river_model_from_X(X, fun_control)
            HoeffdingAdaptiveTreeRegressor()
    """
    core_model = get_one_core_model_from_X(X=X, fun_control=fun_control)
    if fun_control["prep_model"] is not None:
        model = compose.Pipeline(fun_control["prep_model"], core_model)
    else:
        model = core_model
    return model

get_one_sklearn_model_from_X(X, fun_control=None)

Get one sklearn model from X.

Parameters:

Name Type Description Default
X array

The array with the hyper parameter values.

required
fun_control dict

The function control dictionary.

None

Returns:

Type Description
class

The sklearn model.

Examples:

>>> from sklearn.linear_model import LinearRegression
    from spotriver.data.sklearn_hyper_dict import SklearnHyperDict
    fun_control = {}
    add_core_model_to_fun_control(core_model=LinearRegression,
        fun_control=func_control,
        hyper_dict=SklearnHyperDict,
        filename=None)
    X = np.array([0, 0, 0, 0, 0])
    get_one_sklearn_model_from_X(X, fun_control)
    LinearRegression()
Source code in spotpython/hyperparameters/update.py
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
def get_one_sklearn_model_from_X(X, fun_control=None):
    """Get one sklearn model from X.

    Args:
        X (np.array):
            The array with the hyper parameter values.
        fun_control (dict):
            The function control dictionary.

    Returns:
        (class):
            The sklearn model.

    Examples:
        >>> from sklearn.linear_model import LinearRegression
            from spotriver.data.sklearn_hyper_dict import SklearnHyperDict
            fun_control = {}
            add_core_model_to_fun_control(core_model=LinearRegression,
                fun_control=func_control,
                hyper_dict=SklearnHyperDict,
                filename=None)
            X = np.array([0, 0, 0, 0, 0])
            get_one_sklearn_model_from_X(X, fun_control)
            LinearRegression()
    """
    core_model = get_one_core_model_from_X(X=X, fun_control=fun_control)
    if fun_control["prep_model"] is not None:
        model = make_pipeline(fun_control["prep_model"], core_model)
    else:
        model = core_model
    return model

get_transform(fun_control)

Get the transformations of the values from the dictionary fun_control as a list.

Parameters:

Name Type Description Default
fun_control dict

dictionary with levels and types

required

Returns:

Type Description
list

list with transformations

Examples:

>>> from spotpython.hyperparameters.values import get_transform
    d = {"core_model_hyper_dict":{
    "leaf_prediction": {
        "levels": ["mean", "model", "adaptive"],
        "type": "factor",
        "default": "mean",
        "transform": "None",
        "core_model_parameter_type": "str"},
    "leaf_model": {
        "levels": ["linear_model.LinearRegression", "linear_model.PARegressor", "linear_model.Perceptron"],
        "type": "factor",
        "default": "LinearRegression",
        "transform": "None",
        "core_model_parameter_type": "instance"},
    "splitter": {
        "levels": ["EBSTSplitter", "TEBSTSplitter", "QOSplitter"],
        "type": "factor",
        "default": "EBSTSplitter",
        "transform": "None",
        "core_model_parameter_type": "instance()"},
    "binary_split": {
        "levels": [0, 1],
        "type": "factor",
        "default": 0,
        "transform": "None",
        "core_model_parameter_type": "bool"},
    "stop_mem_management": {                                                         "levels": [0, 1],
        "type": "factor",
        "default": 0,
        "transform": "None",
        "core_model_parameter_type": "bool"}}}
get_transform(d)
['None', 'None', 'None', 'None', 'None']
Source code in spotpython/hyperparameters/update.py
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
def get_transform(fun_control) -> list:
    """Get the transformations of the values from the dictionary fun_control as a list.

    Args:
        fun_control (dict):
            dictionary with levels and types

    Returns:
        (list):
            list with transformations

    Examples:
        >>> from spotpython.hyperparameters.values import get_transform
            d = {"core_model_hyper_dict":{
            "leaf_prediction": {
                "levels": ["mean", "model", "adaptive"],
                "type": "factor",
                "default": "mean",
                "transform": "None",
                "core_model_parameter_type": "str"},
            "leaf_model": {
                "levels": ["linear_model.LinearRegression", "linear_model.PARegressor", "linear_model.Perceptron"],
                "type": "factor",
                "default": "LinearRegression",
                "transform": "None",
                "core_model_parameter_type": "instance"},
            "splitter": {
                "levels": ["EBSTSplitter", "TEBSTSplitter", "QOSplitter"],
                "type": "factor",
                "default": "EBSTSplitter",
                "transform": "None",
                "core_model_parameter_type": "instance()"},
            "binary_split": {
                "levels": [0, 1],
                "type": "factor",
                "default": 0,
                "transform": "None",
                "core_model_parameter_type": "bool"},
            "stop_mem_management": {                                                         "levels": [0, 1],
                "type": "factor",
                "default": 0,
                "transform": "None",
                "core_model_parameter_type": "bool"}}}

            get_transform(d)
            ['None', 'None', 'None', 'None', 'None']
    """
    return list(fun_control["core_model_hyper_dict"][key]["transform"] for key in fun_control["core_model_hyper_dict"].keys())

get_tuned_architecture(spot_tuner, fun_control, force_minX=False)

Returns the tuned architecture. If the spot tuner has noise, it returns the architecture with the lowest mean (.min_mean_X), otherwise it returns the architecture with the lowest value (.min_X).

Parameters:

Name Type Description Default
spot_tuner object

spot tuner object.

required
fun_control dict

dictionary containing control parameters for the hyperparameter tuning.

required
force_minX bool

If True, return the architecture with the lowest value (.min_X).

False

Returns:

Type Description
dict

dictionary containing the tuned architecture.

Source code in spotpython/hyperparameters/update.py
 995
 996
 997
 998
 999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
def get_tuned_architecture(spot_tuner, fun_control, force_minX=False) -> dict:
    """
    Returns the tuned architecture. If the spot tuner has noise,
    it returns the architecture with the lowest mean (.min_mean_X),
    otherwise it returns the architecture with the lowest value (.min_X).

    Args:
        spot_tuner (object):
            spot tuner object.
        fun_control (dict):
            dictionary containing control parameters for the hyperparameter tuning.
        force_minX (bool):
            If True, return the architecture with the lowest value (.min_X).

    Returns:
        (dict):
            dictionary containing the tuned architecture.
    """
    if not spot_tuner.noise or force_minX:
        X = spot_tuner.to_all_dim(spot_tuner.min_X.reshape(1, -1))
    else:
        # noise or force_minX is False:
        X = spot_tuner.to_all_dim(spot_tuner.min_mean_X.reshape(1, -1))
    config = get_one_config_from_X(X, fun_control)
    return config

get_tuned_hyperparameters(spot_tuner, fun_control=None)

Get the tuned hyperparameters from the spot tuner. This is just a wrapper function for the spot get_tuned_hyperparameters method.

Parameters:

Name Type Description Default
spot_tuner object

spot tuner object.

required
fun_control dict

dictionary containing control parameters for the hyperparameter tuning. Optional. Default is None.

None

Returns:

Type Description
dict

dictionary containing the tuned hyperparameters.

Examples:

>>> from spotpython.utils.device import getDevice
    from math import inf
    from spotpython.utils.init import fun_control_init
    import numpy as np
    from spotpython.hyperparameters.values import set_control_key_value
    from spotpython.data.diabetes import Diabetes
    from spotpython.hyperparameters.values import get_tuned_hyperparameters
    MAX_TIME = 1
    FUN_EVALS = 10
    INIT_SIZE = 5
    WORKERS = 0
    PREFIX="037"
    DEVICE = getDevice()
    DEVICES = 1
    TEST_SIZE = 0.4
    TORCH_METRIC = "mean_squared_error"
    dataset = Diabetes()
    fun_control = fun_control_init(
        _L_in=10,
        _L_out=1,
        _torchmetric=TORCH_METRIC,
        PREFIX=PREFIX,
        TENSORBOARD_CLEAN=True,
        data_set=dataset,
        device=DEVICE,
        enable_progress_bar=False,
        fun_evals=FUN_EVALS,
        log_level=50,
        max_time=MAX_TIME,
        num_workers=WORKERS,
        show_progress=True,
        test_size=TEST_SIZE,
        tolerance_x=np.sqrt(np.spacing(1)),
        )
    from spotpython.light.regression.netlightregression import NetLightRegression
    from spotpython.hyperdict.light_hyper_dict import LightHyperDict
    from spotpython.hyperparameters.values import add_core_model_to_fun_control
    add_core_model_to_fun_control(fun_control=fun_control,
                                core_model=NetLightRegression,
                                hyper_dict=LightHyperDict)
    from spotpython.hyperparameters.values import set_control_hyperparameter_value
    set_control_hyperparameter_value(fun_control, "l1", [7, 8])
    set_control_hyperparameter_value(fun_control, "epochs", [3, 5])
    set_control_hyperparameter_value(fun_control, "batch_size", [4, 5])
    set_control_hyperparameter_value(fun_control, "optimizer", [
                    "Adam",
                    "RAdam",
                ])
    set_control_hyperparameter_value(fun_control, "dropout_prob", [0.01, 0.1])
    set_control_hyperparameter_value(fun_control, "lr_mult", [0.5, 5.0])
    set_control_hyperparameter_value(fun_control, "patience", [2, 3])
    set_control_hyperparameter_value(fun_control, "act_fn",[
                    "ReLU",
                    "LeakyReLU"
                ] )
    from spotpython.utils.init import design_control_init, surrogate_control_init
    design_control = design_control_init(init_size=INIT_SIZE)
    surrogate_control = surrogate_control_init(noise=True,
                                                n_theta=2)
    from spotpython.fun.hyperlight import HyperLight
    fun = HyperLight(log_level=50).fun
    from spotpython.spot import spot
    spot_tuner = spot.Spot(fun=fun,
                        fun_control=fun_control,
                        design_control=design_control,
                        surrogate_control=surrogate_control)
    spot_tuner.run()
    get_tuned_hyperparameters(spot_tuner)
        {'l1': 7.0,
        'epochs': 5.0,
        'batch_size': 4.0,
        'act_fn': 0.0,
        'optimizer': 0.0,
        'dropout_prob': 0.01,
        'lr_mult': 5.0,
        'patience': 3.0,
        'initialization': 1.0}
Source code in spotpython/hyperparameters/update.py
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
def get_tuned_hyperparameters(spot_tuner, fun_control=None) -> dict:
    """
    Get the tuned hyperparameters from the spot tuner.
    This is just a wrapper function for the spot `get_tuned_hyperparameters` method.

    Args:
        spot_tuner (object):
            spot tuner object.
        fun_control (dict):
            dictionary containing control parameters for the hyperparameter tuning.
            Optional. Default is None.

    Returns:
        (dict):
            dictionary containing the tuned hyperparameters.

    Examples:
        >>> from spotpython.utils.device import getDevice
            from math import inf
            from spotpython.utils.init import fun_control_init
            import numpy as np
            from spotpython.hyperparameters.values import set_control_key_value
            from spotpython.data.diabetes import Diabetes
            from spotpython.hyperparameters.values import get_tuned_hyperparameters
            MAX_TIME = 1
            FUN_EVALS = 10
            INIT_SIZE = 5
            WORKERS = 0
            PREFIX="037"
            DEVICE = getDevice()
            DEVICES = 1
            TEST_SIZE = 0.4
            TORCH_METRIC = "mean_squared_error"
            dataset = Diabetes()
            fun_control = fun_control_init(
                _L_in=10,
                _L_out=1,
                _torchmetric=TORCH_METRIC,
                PREFIX=PREFIX,
                TENSORBOARD_CLEAN=True,
                data_set=dataset,
                device=DEVICE,
                enable_progress_bar=False,
                fun_evals=FUN_EVALS,
                log_level=50,
                max_time=MAX_TIME,
                num_workers=WORKERS,
                show_progress=True,
                test_size=TEST_SIZE,
                tolerance_x=np.sqrt(np.spacing(1)),
                )
            from spotpython.light.regression.netlightregression import NetLightRegression
            from spotpython.hyperdict.light_hyper_dict import LightHyperDict
            from spotpython.hyperparameters.values import add_core_model_to_fun_control
            add_core_model_to_fun_control(fun_control=fun_control,
                                        core_model=NetLightRegression,
                                        hyper_dict=LightHyperDict)
            from spotpython.hyperparameters.values import set_control_hyperparameter_value
            set_control_hyperparameter_value(fun_control, "l1", [7, 8])
            set_control_hyperparameter_value(fun_control, "epochs", [3, 5])
            set_control_hyperparameter_value(fun_control, "batch_size", [4, 5])
            set_control_hyperparameter_value(fun_control, "optimizer", [
                            "Adam",
                            "RAdam",
                        ])
            set_control_hyperparameter_value(fun_control, "dropout_prob", [0.01, 0.1])
            set_control_hyperparameter_value(fun_control, "lr_mult", [0.5, 5.0])
            set_control_hyperparameter_value(fun_control, "patience", [2, 3])
            set_control_hyperparameter_value(fun_control, "act_fn",[
                            "ReLU",
                            "LeakyReLU"
                        ] )
            from spotpython.utils.init import design_control_init, surrogate_control_init
            design_control = design_control_init(init_size=INIT_SIZE)
            surrogate_control = surrogate_control_init(noise=True,
                                                        n_theta=2)
            from spotpython.fun.hyperlight import HyperLight
            fun = HyperLight(log_level=50).fun
            from spotpython.spot import spot
            spot_tuner = spot.Spot(fun=fun,
                                fun_control=fun_control,
                                design_control=design_control,
                                surrogate_control=surrogate_control)
            spot_tuner.run()
            get_tuned_hyperparameters(spot_tuner)
                {'l1': 7.0,
                'epochs': 5.0,
                'batch_size': 4.0,
                'act_fn': 0.0,
                'optimizer': 0.0,
                'dropout_prob': 0.01,
                'lr_mult': 5.0,
                'patience': 3.0,
                'initialization': 1.0}
    """
    return spot_tuner.get_tuned_hyperparameters(fun_control=fun_control)

get_values_from_dict(dictionary)

Get the values from a dictionary as an array. Generate an np.array that contains the values of the keys of a dictionary in the same order as the keys of the dictionary.

Parameters:

Name Type Description Default
dictionary dict

dictionary with values

required

Returns:

Type Description
array

array with values

Examples:

>>> from spotpython.hyperparameters.values import get_values_from_dict
>>> d = {"a": 1, "b": 2, "c": 3}
>>> get_values_from_dict(d)
array([1, 2, 3])
Source code in spotpython/hyperparameters/update.py
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
def get_values_from_dict(dictionary) -> np.array:
    """Get the values from a dictionary as an array.
    Generate an np.array that contains the values of the keys of a dictionary
    in the same order as the keys of the dictionary.

    Args:
        dictionary (dict):
            dictionary with values

    Returns:
        (np.array):
            array with values

    Examples:
        >>> from spotpython.hyperparameters.values import get_values_from_dict
        >>> d = {"a": 1, "b": 2, "c": 3}
        >>> get_values_from_dict(d)
        array([1, 2, 3])
    """
    return np.array(list(dictionary.values()))

get_var_name(fun_control)

Get the names of the values from the dictionary fun_control as a list. If no “core_model_hyper_dict” key exists in fun_control, return None.

Parameters:

Name Type Description Default
fun_control dict

dictionary with names

required

Returns:

Type Description
list

ist with names

Examples:

>>> from spotpython.hyperparameters.values import get_var_name
    fun_control = {"core_model_hyper_dict":{
                "leaf_prediction": {
                    "levels": ["mean", "model", "adaptive"],
                    "type": "factor",
                    "default": "mean",
                    "core_model_parameter_type": "str"},
                "leaf_model": {
                    "levels": ["linear_model.LinearRegression",
                                "linear_model.PARegressor",
                                "linear_model.Perceptron"],
                    "type": "factor",
                    "default": "LinearRegression",
                    "core_model_parameter_type": "instance"},
                "splitter": {
                    "levels": ["EBSTSplitter", "TEBSTSplitter", "QOSplitter"],
                    "type": "factor",
                    "default": "EBSTSplitter",
                    "core_model_parameter_type": "instance()"},
                "binary_split": {
                    "levels": [0, 1],
                    "type": "factor",
                    "default": 0,
                    "core_model_parameter_type": "bool"},
                "stop_mem_management": {
                    "levels": [0, 1],
                    "type": "factor",
                    "default": 0,
                    "core_model_parameter_type": "bool"}}}
    get_var_name(fun_control)
    ['leaf_prediction',
        'leaf_model',
        'splitter',
        'binary_split',
        'stop_mem_management']
Source code in spotpython/hyperparameters/update.py
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
def get_var_name(fun_control) -> list:
    """Get the names of the values from the dictionary fun_control as a list.
    If no "core_model_hyper_dict" key exists in fun_control, return None.

    Args:
        fun_control (dict):
            dictionary with names

    Returns:
        (list):
            ist with names

    Examples:
        >>> from spotpython.hyperparameters.values import get_var_name
            fun_control = {"core_model_hyper_dict":{
                        "leaf_prediction": {
                            "levels": ["mean", "model", "adaptive"],
                            "type": "factor",
                            "default": "mean",
                            "core_model_parameter_type": "str"},
                        "leaf_model": {
                            "levels": ["linear_model.LinearRegression",
                                        "linear_model.PARegressor",
                                        "linear_model.Perceptron"],
                            "type": "factor",
                            "default": "LinearRegression",
                            "core_model_parameter_type": "instance"},
                        "splitter": {
                            "levels": ["EBSTSplitter", "TEBSTSplitter", "QOSplitter"],
                            "type": "factor",
                            "default": "EBSTSplitter",
                            "core_model_parameter_type": "instance()"},
                        "binary_split": {
                            "levels": [0, 1],
                            "type": "factor",
                            "default": 0,
                            "core_model_parameter_type": "bool"},
                        "stop_mem_management": {
                            "levels": [0, 1],
                            "type": "factor",
                            "default": 0,
                            "core_model_parameter_type": "bool"}}}
            get_var_name(fun_control)
            ['leaf_prediction',
                'leaf_model',
                'splitter',
                'binary_split',
                'stop_mem_management']
    """
    if "core_model_hyper_dict" not in fun_control.keys():
        return None
    else:
        return list(fun_control["core_model_hyper_dict"].keys())

get_var_type(fun_control)

Get the types of the values from the dictionary fun_control as a list. If no “core_model_hyper_dict” key exists in fun_control, return None.

Parameters:

Name Type Description Default
fun_control dict

dictionary with levels and types

required

Returns:

Type Description
list

list with types

Examples:

>>> from spotpython.hyperparameters.values import get_var_type
    d = {"core_model_hyper_dict":{
    "leaf_prediction": {
        "levels": ["mean", "model", "adaptive"],
        "type": "factor",
        "default": "mean",
        "core_model_parameter_type": "str"},
    "leaf_model": {
        "levels": ["linear_model.LinearRegression", "linear_model.PARegressor", "linear_model.Perceptron"],
        "type": "factor",
        "default": "LinearRegression",
        "core_model_parameter_type": "instance"},
    "splitter": {
        "levels": ["EBSTSplitter", "TEBSTSplitter", "QOSplitter"],
        "type": "factor",
        "default": "EBSTSplitter",
        "core_model_parameter_type": "instance()"},
    "binary_split": {
        "levels": [0, 1],
        "type": "factor",
        "default": 0,
        "core_model_parameter_type": "bool"},
    "stop_mem_management": {                                                         "levels": [0, 1],
        "type": "factor",
        "default": 0,
        "core_model_parameter_type": "bool"}}}
get_var_type(d)
['factor', 'factor', 'factor', 'factor', 'factor']
Source code in spotpython/hyperparameters/update.py
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
def get_var_type(fun_control) -> list:
    """
    Get the types of the values from the dictionary fun_control as a list.
    If no "core_model_hyper_dict" key exists in fun_control, return None.

    Args:
        fun_control (dict):
            dictionary with levels and types

    Returns:
        (list):
            list with types

    Examples:
        >>> from spotpython.hyperparameters.values import get_var_type
            d = {"core_model_hyper_dict":{
            "leaf_prediction": {
                "levels": ["mean", "model", "adaptive"],
                "type": "factor",
                "default": "mean",
                "core_model_parameter_type": "str"},
            "leaf_model": {
                "levels": ["linear_model.LinearRegression", "linear_model.PARegressor", "linear_model.Perceptron"],
                "type": "factor",
                "default": "LinearRegression",
                "core_model_parameter_type": "instance"},
            "splitter": {
                "levels": ["EBSTSplitter", "TEBSTSplitter", "QOSplitter"],
                "type": "factor",
                "default": "EBSTSplitter",
                "core_model_parameter_type": "instance()"},
            "binary_split": {
                "levels": [0, 1],
                "type": "factor",
                "default": 0,
                "core_model_parameter_type": "bool"},
            "stop_mem_management": {                                                         "levels": [0, 1],
                "type": "factor",
                "default": 0,
                "core_model_parameter_type": "bool"}}}

            get_var_type(d)
            ['factor', 'factor', 'factor', 'factor', 'factor']
    """
    if "core_model_hyper_dict" not in fun_control.keys():
        return None
    else:
        return list(fun_control["core_model_hyper_dict"][key]["type"] for key in fun_control["core_model_hyper_dict"].keys())

get_var_type_from_var_name(fun_control, var_name)

This function gets the variable type from the variable name.

Parameters:

Name Type Description Default
fun_control dict

fun_control dictionary

required
var_name str

variable name

required

Returns:

Type Description
str

variable type

Examples:

>>> from spotpython.utils.init import fun_control_init
    from spotpython.hyperparameters.values import get_var_type_from_var_name
    control_dict = fun_control_init()
    get_var_type_from_var_name(var_name="max_depth",
                    fun_control=control_dict)
    "int"
Source code in spotpython/hyperparameters/update.py
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
def get_var_type_from_var_name(fun_control, var_name) -> str:
    """
    This function gets the variable type from the variable name.

    Args:
        fun_control (dict): fun_control dictionary
        var_name (str): variable name

    Returns:
        (str): variable type

    Examples:
        >>> from spotpython.utils.init import fun_control_init
            from spotpython.hyperparameters.values import get_var_type_from_var_name
            control_dict = fun_control_init()
            get_var_type_from_var_name(var_name="max_depth",
                            fun_control=control_dict)
            "int"
    """
    var_type_list = get_control_key_value(control_dict=fun_control, key="var_type")
    var_name_list = get_control_key_value(control_dict=fun_control, key="var_name")
    return var_type_list[var_name_list.index(var_name)]

iterate_dict_values(var_dict)

Iterate over the values of a dictionary of variables. This function takes a dictionary of variables as input arguments and returns a generator that yields dictionaries with the values from the arrays in the input dictionary.

Parameters:

Name Type Description Default
var_dict dict

A dictionary where keys are variable names and values are numpy arrays.

required

Returns:

Type Description
Generator[Dict[str, Union[int, float]], None, None]

Generator[dict]: A generator that yields dictionaries with the values from the arrays in the input dictionary.

Examples:

>>> import numpy as np
>>> from spotpython.hyperparameters.values import iterate_dict_values
>>> var_dict = {'a': np.array([1, 3, 5]), 'b': np.array([2, 4, 6])}
>>> list(iterate_dict_values(var_dict))
[{'a': 1, 'b': 2}, {'a': 3, 'b': 4}, {'a': 5, 'b': 6}]
Source code in spotpython/hyperparameters/update.py
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
def iterate_dict_values(var_dict: Dict[str, np.ndarray]) -> Generator[Dict[str, Union[int, float]], None, None]:
    """Iterate over the values of a dictionary of variables.
    This function takes a dictionary of variables as input arguments and returns a generator that
    yields dictionaries with the values from the arrays in the input dictionary.

    Args:
        var_dict (dict): A dictionary where keys are variable names and values are numpy arrays.

    Returns:
        Generator[dict]:
            A generator that yields dictionaries with the values from the arrays in the input dictionary.

    Examples:
        >>> import numpy as np
        >>> from spotpython.hyperparameters.values import iterate_dict_values
        >>> var_dict = {'a': np.array([1, 3, 5]), 'b': np.array([2, 4, 6])}
        >>> list(iterate_dict_values(var_dict))
        [{'a': 1, 'b': 2}, {'a': 3, 'b': 4}, {'a': 5, 'b': 6}]
    """
    n = len(next(iter(var_dict.values())))
    for i in range(n):
        yield {key: value[i] for key, value in var_dict.items()}

modify_boolean_hyper_parameter_levels(fun_control, hyperparameter, levels)

This function modifies the levels of a boolean hyperparameter in the fun_control dictionary. It also sets the lower and upper bounds of the hyperparameter to 0 and len(levels) - 1, respectively.

Parameters:

Name Type Description Default
fun_control dict

fun_control dictionary

required
hyperparameter str

hyperparameter name

required
levels list

list of levels

required

Returns:

Type Description
None

None.

Source code in spotpython/hyperparameters/update.py
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
def modify_boolean_hyper_parameter_levels(fun_control, hyperparameter, levels) -> None:
    """
    This function modifies the levels of a boolean hyperparameter in the fun_control dictionary.
    It also sets the lower and upper bounds of the hyperparameter to 0 and len(levels) - 1, respectively.

    Args:
        fun_control (dict):
            fun_control dictionary
        hyperparameter (str):
            hyperparameter name
        levels (list):
            list of levels

    Returns:
        None.
    """
    fun_control["core_model_hyper_dict"][hyperparameter].update({"levels": levels})
    fun_control["core_model_hyper_dict"][hyperparameter].update({"lower": levels[0]})
    fun_control["core_model_hyper_dict"][hyperparameter].update({"upper": levels[1]})

modify_hyper_parameter_bounds(fun_control, hyperparameter, bounds)

Modify the bounds of a hyperparameter in the fun_control dictionary.

Parameters:

Name Type Description Default
fun_control dict

fun_control dictionary

required
hyperparameter str

hyperparameter name

required
bounds list

list of two bound values. The first value represents the lower bound and the second value represents the upper bound.

required

Returns:

Type Description
None

None.

Examples:

>>> from spotpython.hyperparameters.values import modify_hyper_parameter_levels
    fun_control = {}
    core_model  = HoeffdingTreeRegressor
    fun_control.update({"core_model": core_model})
    fun_control.update({"core_model_hyper_dict": river_hyper_dict[core_model.__name__]})
    bounds = [3, 11]
    fun_control = modify_hyper_parameter_levels(fun_control, "min_samples_split", bounds)
Source code in spotpython/hyperparameters/update.py
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
def modify_hyper_parameter_bounds(fun_control, hyperparameter, bounds) -> None:
    """
    Modify the bounds of a hyperparameter in the fun_control dictionary.

    Args:
        fun_control (dict):
            fun_control dictionary
        hyperparameter (str):
            hyperparameter name
        bounds (list):
            list of two bound values. The first value represents the lower bound
            and the second value represents the upper bound.

    Returns:
        None.

    Examples:
        >>> from spotpython.hyperparameters.values import modify_hyper_parameter_levels
            fun_control = {}
            core_model  = HoeffdingTreeRegressor
            fun_control.update({"core_model": core_model})
            fun_control.update({"core_model_hyper_dict": river_hyper_dict[core_model.__name__]})
            bounds = [3, 11]
            fun_control = modify_hyper_parameter_levels(fun_control, "min_samples_split", bounds)
    """
    fun_control["core_model_hyper_dict"][hyperparameter].update({"lower": bounds[0]})
    fun_control["core_model_hyper_dict"][hyperparameter].update({"upper": bounds[1]})

modify_hyper_parameter_levels(fun_control, hyperparameter, levels)

This function modifies the levels of a hyperparameter in the fun_control dictionary. It also sets the lower and upper bounds of the hyperparameter to 0 and len(levels) - 1, respectively.

Parameters:

Name Type Description Default
fun_control dict

fun_control dictionary

required
hyperparameter str

hyperparameter name

required
levels list

list of levels

required

Returns:

Type Description
None

None.

Examples:

>>> fun_control = {}
    from spotpython.hyperparameters.values import modify_hyper_parameter_levels
    core_model  = HoeffdingTreeRegressor
    fun_control.update({"core_model": core_model})
    fun_control.update({"core_model_hyper_dict": river_hyper_dict[core_model.__name__]})
    levels = ["mean", "model"]
    fun_control = modify_hyper_parameter_levels(fun_control, "leaf_prediction", levels)
Source code in spotpython/hyperparameters/update.py
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
def modify_hyper_parameter_levels(fun_control, hyperparameter, levels) -> None:
    """
    This function modifies the levels of a hyperparameter in the fun_control dictionary.
    It also sets the lower and upper bounds of the hyperparameter to 0 and len(levels) - 1, respectively.

    Args:
        fun_control (dict):
            fun_control dictionary
        hyperparameter (str):
            hyperparameter name
        levels (list):
            list of levels

    Returns:
        None.

    Examples:
        >>> fun_control = {}
            from spotpython.hyperparameters.values import modify_hyper_parameter_levels
            core_model  = HoeffdingTreeRegressor
            fun_control.update({"core_model": core_model})
            fun_control.update({"core_model_hyper_dict": river_hyper_dict[core_model.__name__]})
            levels = ["mean", "model"]
            fun_control = modify_hyper_parameter_levels(fun_control, "leaf_prediction", levels)
    """
    fun_control["core_model_hyper_dict"][hyperparameter].update({"levels": levels})
    fun_control["core_model_hyper_dict"][hyperparameter].update({"lower": 0})
    fun_control["core_model_hyper_dict"][hyperparameter].update({"upper": len(levels) - 1})

replace_levels_with_positions(hyper_dict, hyper_dict_values)

Replace the levels with the position in the levels list. The function that takes two dictionaries. The first contains as hyperparameters as keys. If the hyperparameter has the key “levels”, then the value of the corresponding hyperparameter in the second dictionary is replaced by the position of the value in the list of levels. The function returns a dictionary with the same keys as the second dictionary. For example, if the second dictionary is {“a”: 1, “b”: “model1”, “c”: 3} and the first dictionary is { “a”: {“type”: “int”}, “b”: {“levels”: [“model4”, “model5”, “model1”]}, “d”: {“type”: “float”}}, then the function should return {“a”: 1, “b”: 2, “c”: 3}.

Parameters:

Name Type Description Default
hyper_dict dict

dictionary with levels

required
hyper_dict_values dict

dictionary with values

required

Returns:

Type Description
dict

dictionary with values

Examples:

>>> from spotpython.hyperparameters.values import replace_levels_with_positions
    hyper_dict = {"leaf_prediction": {
        "levels": ["mean", "model", "adaptive"],
        "type": "factor",
        "default": "mean",
        "core_model_parameter_type": "str"},
        "leaf_model": {
            "levels": ["linear_model.LinearRegression", "linear_model.PARegressor", "linear_model.Perceptron"],
            "type": "factor",
            "default": "LinearRegression",
            "core_model_parameter_type": "instance"},
        "splitter": {
            "levels": ["EBSTSplitter", "TEBSTSplitter", "QOSplitter"],
            "type": "factor",
            "default": "EBSTSplitter",
            "core_model_parameter_type": "instance()"},
        "binary_split": {
            "levels": [0, 1],
            "type": "factor",
            "default": 0,
            "core_model_parameter_type": "bool"},
        "stop_mem_management": {
            "levels": [0, 1],
            "type": "factor",
            "default": 0,
            "core_model_parameter_type": "bool"}}
    hyper_dict_values = {"leaf_prediction": "mean",
        "leaf_model": "linear_model.LinearRegression",
        "splitter": "EBSTSplitter",
        "binary_split": 0,
        "stop_mem_management": 0}
    replace_levels_with_position(hyper_dict, hyper_dict_values)
        {'leaf_prediction': 0,
        'leaf_model': 0,
        'splitter': 0,
        'binary_split': 0,
        'stop_mem_management': 0}
Source code in spotpython/hyperparameters/update.py
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
def replace_levels_with_positions(hyper_dict, hyper_dict_values) -> dict:
    """Replace the levels with the position in the levels list.
    The function that takes two dictionaries.
    The first contains as hyperparameters as keys.
    If the hyperparameter has the key "levels",
    then the value of the corresponding hyperparameter in the second dictionary is
    replaced by the position of the value in the list of levels.
    The function returns a dictionary with the same keys as the second dictionary.
    For example, if the second dictionary is {"a": 1, "b": "model1", "c": 3}
    and the first dictionary is {
        "a": {"type": "int"},
        "b": {"levels": ["model4", "model5", "model1"]},
        "d": {"type": "float"}},
    then the function should return {"a": 1, "b": 2, "c": 3}.

    Args:
        hyper_dict (dict):
            dictionary with levels
        hyper_dict_values (dict):
            dictionary with values

    Returns:
        (dict):
            dictionary with values

    Examples:
        >>> from spotpython.hyperparameters.values import replace_levels_with_positions
            hyper_dict = {"leaf_prediction": {
                "levels": ["mean", "model", "adaptive"],
                "type": "factor",
                "default": "mean",
                "core_model_parameter_type": "str"},
                "leaf_model": {
                    "levels": ["linear_model.LinearRegression", "linear_model.PARegressor", "linear_model.Perceptron"],
                    "type": "factor",
                    "default": "LinearRegression",
                    "core_model_parameter_type": "instance"},
                "splitter": {
                    "levels": ["EBSTSplitter", "TEBSTSplitter", "QOSplitter"],
                    "type": "factor",
                    "default": "EBSTSplitter",
                    "core_model_parameter_type": "instance()"},
                "binary_split": {
                    "levels": [0, 1],
                    "type": "factor",
                    "default": 0,
                    "core_model_parameter_type": "bool"},
                "stop_mem_management": {
                    "levels": [0, 1],
                    "type": "factor",
                    "default": 0,
                    "core_model_parameter_type": "bool"}}
            hyper_dict_values = {"leaf_prediction": "mean",
                "leaf_model": "linear_model.LinearRegression",
                "splitter": "EBSTSplitter",
                "binary_split": 0,
                "stop_mem_management": 0}
            replace_levels_with_position(hyper_dict, hyper_dict_values)
                {'leaf_prediction': 0,
                'leaf_model': 0,
                'splitter': 0,
                'binary_split': 0,
                'stop_mem_management': 0}
    """
    hyper_dict_values_new = copy.deepcopy(hyper_dict_values)
    # generate an error if the following code fails and write an error message:
    try:
        for key, value in hyper_dict_values.items():
            if key in hyper_dict.keys():
                if "levels" in hyper_dict[key].keys():
                    hyper_dict_values_new[key] = hyper_dict[key]["levels"].index(value)
    except Exception as e:
        print("!!! Warning: ", e)
        print("Did you modify lower and upper bounds so that the default values are not included?")
        print("Returning 'None'.")
        return None
    return hyper_dict_values_new

return_conf_list_from_var_dict(var_dict, fun_control, default=False)

Return a list of configurations from a dictionary of variables.

This function takes a dictionary of variables and a dictionary of function control as input arguments. It performs similar steps as generate_one_config_from_var_dict() but returns a list of dictionaries of hyper parameter values.

Parameters:

Name Type Description Default
var_dict dict

A dictionary where keys are variable names and values are numpy arrays.

required
fun_control dict

A dictionary which (at least) has an entry with the following key: “var_type” (list): A list of variable types. If the entry is not “num” the corresponding value will be converted to the type “int”.

required

Returns:

Name Type Description
list List[Dict[str, Union[int, float]]]

A list of dictionaries of hyper parameter values. Transformations are applied to the values.

Examples:

>>> import numpy as np
>>> from spotpython.hyperparameters.values import return_conf_list_from_var_dict
>>> var_dict = {'a': np.array([1, 3, 5]), 'b': np.array([2, 4, 6])}
>>> fun_control = {'var_type': ['int', 'int']}
>>> return_conf_list_from_var_dict(var_dict, fun_control)
[{'a': 1, 'b': 2}, {'a': 3, 'b': 4}, {'a': 5, 'b': 6}]
Source code in spotpython/hyperparameters/update.py
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
def return_conf_list_from_var_dict(
    var_dict: Dict[str, np.ndarray],
    fun_control: Dict[str, Union[List[str], str]],
    default: bool = False,
) -> List[Dict[str, Union[int, float]]]:
    """Return a list of configurations from a dictionary of variables.

    This function takes a dictionary of variables and a dictionary of function control as input arguments.
    It performs similar steps as generate_one_config_from_var_dict() but returns a list of dictionaries
    of hyper parameter values.

    Args:
        var_dict (dict): A dictionary where keys are variable names and values are numpy arrays.
        fun_control (dict): A dictionary which (at least) has an entry with the following key:
            "var_type" (list): A list of variable types. If the entry is not "num" the corresponding
            value will be converted to the type "int".

    Returns:
        list: A list of dictionaries of hyper parameter values. Transformations are applied to the values.

    Examples:
        >>> import numpy as np
        >>> from spotpython.hyperparameters.values import return_conf_list_from_var_dict
        >>> var_dict = {'a': np.array([1, 3, 5]), 'b': np.array([2, 4, 6])}
        >>> fun_control = {'var_type': ['int', 'int']}
        >>> return_conf_list_from_var_dict(var_dict, fun_control)
        [{'a': 1, 'b': 2}, {'a': 3, 'b': 4}, {'a': 5, 'b': 6}]
    """
    conf_list = []
    for values in generate_one_config_from_var_dict(var_dict, fun_control, default=default):
        conf_list.append(values)
    return conf_list

set_control_hyperparameter_value(control_dict, hyperparameter, value)

This function sets the hyperparameter values depending on the var_type via modify_hyperameter_levels or modify_hyperparameter_bounds in the control_dict dictionary. If the hyperparameter is a factor, it calls modify_hyper_parameter_levels. Otherwise, it calls modify_hyper_parameter_bounds.

Parameters:

Name Type Description Default
control_dict dict

control_dict dictionary

required
hyperparameter str

key

required
value Any

value

required

Returns:

Type Description
None

None.

Source code in spotpython/hyperparameters/update.py
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
def set_control_hyperparameter_value(control_dict, hyperparameter, value) -> None:
    """
    This function sets the hyperparameter values depending on the var_type
    via modify_hyperameter_levels or modify_hyperparameter_bounds in the control_dict dictionary.
    If the hyperparameter is a factor, it calls modify_hyper_parameter_levels.
    Otherwise, it calls modify_hyper_parameter_bounds.

    Args:
        control_dict (dict):
            control_dict dictionary
        hyperparameter (str): key
        value (Any): value

    Returns:
        None.

    """
    print(f"Setting hyperparameter {hyperparameter} to value {value}.")
    vt = get_var_type_from_var_name(fun_control=control_dict, var_name=hyperparameter)
    print(f"Variable type is {vt}.")
    core_type = get_core_model_parameter_type_from_var_name(fun_control=control_dict, var_name=hyperparameter)
    print(f"Core type is {core_type}.")
    if vt == "factor" and core_type != "bool":
        print("Calling modify_hyper_parameter_levels().")
        modify_hyper_parameter_levels(fun_control=control_dict, hyperparameter=hyperparameter, levels=value)
    elif vt == "factor" and core_type == "bool":
        print("Calling modify_boolean_hyper_parameter_levels().")
        modify_boolean_hyper_parameter_levels(fun_control=control_dict, hyperparameter=hyperparameter, levels=value)
    else:
        print("Calling modify_hyper_parameter_bounds().")
        modify_hyper_parameter_bounds(fun_control=control_dict, hyperparameter=hyperparameter, bounds=value)

set_control_key_value(control_dict, key, value, replace=False)

This function sets the key value pair in the control_dict dictionary.

Parameters:

Name Type Description Default
control_dict dict

control_dict dictionary

required
key str

key

required
value Any

value

required
replace bool

replace value if key already exists. Default is False.

False

Returns:

Type Description
None

None.

Attributes:

Name Type Description
key str

key

value Any

value

Examples:

>>> from spotpython.utils.init import fun_control_init
    from spotpython.hyperparameters.values import set_control_key_value
    control_dict = fun_control_init()
    set_control_key_value(control_dict=control_dict,
                  key="key",
                  value="value")
    control_dict["key"]
Source code in spotpython/hyperparameters/update.py
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
def set_control_key_value(control_dict, key, value, replace=False) -> None:
    """
    This function sets the key value pair in the control_dict dictionary.

    Args:
        control_dict (dict):
            control_dict dictionary
        key (str): key
        value (Any): value
        replace (bool): replace value if key already exists. Default is False.

    Returns:
        None.

    Attributes:
        key (str): key
        value (Any): value

    Examples:
        >>> from spotpython.utils.init import fun_control_init
            from spotpython.hyperparameters.values import set_control_key_value
            control_dict = fun_control_init()
            set_control_key_value(control_dict=control_dict,
                          key="key",
                          value="value")
            control_dict["key"]

    """
    if replace:
        control_dict.update({key: value})
    else:
        if key not in control_dict.keys():
            control_dict.update({key: value})

update_fun_control_with_hyper_num_cat_dicts(fun_control, num_dict, cat_dict, dict)

Update an existing fun_control dictionary with new hyperparameter values. All values from the hyperparameter dict (dict) are updated in the fun_control dictionary using the num_dict and cat_dict dictionaries.

Parameters:

Name Type Description Default
fun_control dict

The fun_control dictionary. This dictionary is updated with the new hyperparameter values.

required
num_dict dict

The dictionary containing the numerical hyperparameter values, which are used to update the fun_control dictionary.

required
cat_dict dict

The dictionary containing the categorical hyperparameter values, which are used to update the fun_control dictionary.

required
dict dict

The dictionary containing the “old” hyperparameter values.

required
Source code in spotpython/hyperparameters/update.py
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
def update_fun_control_with_hyper_num_cat_dicts(fun_control, num_dict, cat_dict, dict):
    """
    Update an existing fun_control dictionary with new hyperparameter values.
    All values from the hyperparameter dict (dict) are updated in the fun_control dictionary
    using the num_dict and cat_dict dictionaries.

    Args:
        fun_control (dict):
            The fun_control dictionary. This dictionary is updated with the new hyperparameter values.
        num_dict (dict):
            The dictionary containing the numerical hyperparameter values, which
            are used to update the fun_control dictionary.
        cat_dict (dict):
            The dictionary containing the categorical hyperparameter values, which
            are used to update the fun_control dictionary.
        dict (dict):
            The dictionary containing the "old" hyperparameter values.
    """
    for i, (key, value) in enumerate(dict.items()):
        if dict[key]["type"] == "int":
            set_control_hyperparameter_value(
                fun_control,
                key,
                [
                    int(num_dict[key]["lower"]),
                    int(num_dict[key]["upper"]),
                ],
            )
        if (dict[key]["type"] == "factor") and (dict[key]["core_model_parameter_type"] == "bool"):
            set_control_hyperparameter_value(
                fun_control,
                key,
                [
                    int(num_dict[key]["lower"]),
                    int(num_dict[key]["upper"]),
                ],
            )
        if dict[key]["type"] == "float":
            set_control_hyperparameter_value(
                fun_control,
                key,
                [
                    float(num_dict[key]["lower"]),
                    float(num_dict[key]["upper"]),
                ],
            )
        if dict[key]["type"] == "factor" and dict[key]["core_model_parameter_type"] != "bool":
            fle = cat_dict[key]["levels"]
            # convert the string to a list of strings
            fle = fle.split()
            set_control_hyperparameter_value(fun_control, key, fle)
            fun_control["core_model_hyper_dict"][key].update({"upper": len(fle) - 1})