skpro.metrics.survival.ConcordanceHarrell#
- class skpro.metrics.survival.ConcordanceHarrell(score='mean', score_args=None, higher_score_is_lower_risk=True, tie_score=0.5, normalization='overall', multioutput='uniform_average', multivariate=False)[source]#
Concordance index (Harrell).
Fraction of concordant test index pairs among all comparale pairs, as proposed in [1], commonly known as Harrell’s C-index, Harrell’s C, or simply concordance index, if not in delination of other C-indices (e.g., Uno’s C-index).
For ground truth samples \(y_i, c_i, i = 1 \dots N\), and predicted inverse risk scores \(s_i, i = 1 \dots N\), a pair of test non-equal test indices \(i \lneq j\) is concordant if \((y_i > y_j) \land (s_i > s_j)\) or \((y_i < y_j) \land (s_i < s_j)\). If \((s_i = s_j)\), the pair is counted as concordant if \(y_i = y_j\), and \(c_i = c_j = 0\), otherwise it is considered a tie, counted as half concordant, half discordant by default.
A pair of test indices \(i \lneq j\) is said to be comparable if one of the following conditions holds:
\(y_i > y_j\) and \(c_j = 0\)
\(y_i < y_j\) and \(c_i = 0\)
\(y_i = y_j\) and \(c_i c_j = 0\)
This metric supports multiple options for inverse risk scores, including any method evaluates of predictive distributions.
The default is the predictive mean survival time.
evaluate computes the concordance index. evaluate_by_index produces, for one test sample, the fraction of concordant pairs among all pairs with this sample as first index. multivariate controls averaging over variables.
- Parameters:
- scorestr, optional, default=’mean’
The type of inverse risk score to use. Calls predict_proba, then the method of the same name as score. Examples include ‘mean’, ‘median’, ‘quantile’, ‘cdf’.
- score_argsdict, optional, default=None
Additional arguments to pass to the score method, e.g., quantiles.
- higher_score_is_lower_riskbool, optional, default=True
If True, higher score is considered lower risk, and vice versa, that is, the score is assumed to be an inverse risk score. If False, the score is assumed to be a risk score, and a negative sign is applied to the score.
- tie_scorefloat, optional, default=0.5
The value to use for ties in the risk scores, as a relative value to counting as concordant. 1 is counting as concordant, 0 is counting as discordant. 0.5 is counting as half concordant, half discordant.
- normalizationstr, {‘overall’, ‘index’}, optional, default=’overall’
Determines the normalization of the concordance index, whether fractions of concordant pairs are averaged primarily overall, or primarily per index. In both cases,
evaluate
returns the arithmetic mean ofevaluate_by_index
.If
'overall'
,evaluate
returns the fraction of concordant among all comparable pairs. This is as in [1]. In
evaluate_by_index
, fraction denominators are the number of comparable pairs overall, divided by the number of samples, instead of the number of comparable pairs in which the index is the first index.If
'index'
,evaluate
returns the average, over indices,
of the fraction of concordant pairs among all comparable pairs in which the index is the first index. In
evaluate_by_index
, entries are the fraction of concordant pairs among all comparable pairs in which the index is the first index, without further normalization.- multioutput{‘raw_values’, ‘uniform_average’} or array-like of shape
(n_outputs,), default=’uniform_average’ Defines whether and how to aggregate metric for across variables. If ‘uniform_average’ (default), errors are mean-averaged across variables. If array-like, errors are weighted averaged across variables, values as weights. If ‘raw_values’, does not average errors across variables, columns are retained.
- multivariatebool, optional, default=False
if True, behaves as multivariate log-loss C-index is computed for entire row, results one score per row if False, is univariate log-loss C-index is computed per variable marginal, results in many scores per row
References
- Attributes:
name
Return the name of the object or estimator.
Methods
__call__
(y_true, y_pred, **kwargs)Calculate metric value using underlying metric function.
clone
()Obtain a clone of the object with same hyper-parameters.
clone_tags
(estimator[, tag_names])Clone tags from another estimator as dynamic override.
create_test_instance
([parameter_set])Construct Estimator instance if possible.
create_test_instances_and_names
([parameter_set])Create list of all test instances and a list of names for them.
evaluate
(y_true, y_pred, **kwargs)Evaluate the metric on given inputs.
evaluate_by_index
(y_true, y_pred, **kwargs)Evaluate the metric by instance index (row).
get_class_tag
(tag_name[, tag_value_default])Get a class tag's value.
Get class tags from the class and all its parent classes.
Get config flags for self.
Get object's parameter defaults.
Get object's parameter names.
get_params
([deep])Get a dict of parameters values for this object.
get_tag
(tag_name[, tag_value_default, ...])Get tag value from estimator class and dynamic tag overrides.
get_tags
()Get tags from estimator class and dynamic tag overrides.
get_test_params
([parameter_set])Return testing parameter settings for the estimator.
Check if the object is composed of other BaseObjects.
reset
()Reset the object to a clean post-init state.
set_config
(**config_dict)Set config flags to given values.
set_params
(**params)Set the parameters of this object.
set_random_state
([random_state, deep, ...])Set random_state pseudo-random seed parameters for self.
set_tags
(**tag_dict)Set dynamic tags to given values.
- clone()[source]#
Obtain a clone of the object with same hyper-parameters.
A clone is a different object without shared references, in post-init state. This function is equivalent to returning sklearn.clone of self.
- Raises:
- RuntimeError if the clone is non-conforming, due to faulty
__init__
.
- RuntimeError if the clone is non-conforming, due to faulty
Notes
If successful, equal in value to
type(self)(**self.get_params(deep=False))
.
- clone_tags(estimator, tag_names=None)[source]#
Clone tags from another estimator as dynamic override.
- Parameters:
- estimatorestimator inheriting from :class:BaseEstimator
- tag_namesstr or list of str, default = None
Names of tags to clone. If None then all tags in estimator are used as tag_names.
- Returns:
- Self
Reference to self.
Notes
Changes object state by setting tag values in tag_set from estimator as dynamic tags in self.
- classmethod create_test_instance(parameter_set='default')[source]#
Construct Estimator instance if possible.
- Parameters:
- parameter_setstr, default=”default”
Name of the set of test parameters to return, for use in tests. If no special parameters are defined for a value, will return “default” set.
- Returns:
- instanceinstance of the class with default parameters
Notes
get_test_params can return dict or list of dict. This function takes first or single dict that get_test_params returns, and constructs the object with that.
- classmethod create_test_instances_and_names(parameter_set='default')[source]#
Create list of all test instances and a list of names for them.
- Parameters:
- parameter_setstr, default=”default”
Name of the set of test parameters to return, for use in tests. If no special parameters are defined for a value, will return “default” set.
- Returns:
- objslist of instances of cls
i-th instance is cls(**cls.get_test_params()[i])
- nameslist of str, same length as objs
i-th element is name of i-th instance of obj in tests convention is {cls.__name__}-{i} if more than one instance otherwise {cls.__name__}
- evaluate(y_true, y_pred, **kwargs)[source]#
Evaluate the metric on given inputs.
- Parameters:
- y_truepd.Series, pd.DataFrame, 1D np.array, or 2D np.ndarray
Ground truth (correct) target values.
- y_predreturn object of probabilistic predictition method scitype:y_pred
must have same index and columns as y_true Predicted values, i-th row is prediction for i-th row of
y_true
.
- Returns:
- lossfloat or 1-column pd.DataFrame with calculated metric value(s)
float if multioutput = “uniform_average” or multivariate = True 1-column df if multioutput = “raw_values” and metric is not multivariate metric is always averaged (arithmetic) over rows
- evaluate_by_index(y_true, y_pred, **kwargs)[source]#
Evaluate the metric by instance index (row).
- Parameters:
- y_truepd.Series, pd.DataFrame, 1D np.array, or 2D np.ndarray
Ground truth (correct) target values.
- y_predskpro BaseDistribution of same shape as y_true
Predictive distribution. Must have same index and columns as y_true.
- classmethod get_class_tag(tag_name, tag_value_default=None)[source]#
Get a class tag’s value.
Does not return information from dynamic tags (set via set_tags or clone_tags) that are defined on instances.
- Parameters:
- tag_namestr
Name of tag value.
- tag_value_defaultany
Default/fallback value if tag is not found.
- Returns:
- tag_value
Value of the tag_name tag in self. If not found, returns tag_value_default.
- classmethod get_class_tags()[source]#
Get class tags from the class and all its parent classes.
Retrieves tag: value pairs from _tags class attribute. Does not return information from dynamic tags (set via set_tags or clone_tags) that are defined on instances.
- Returns:
- collected_tagsdict
Dictionary of class tag name: tag value pairs. Collected from _tags class attribute via nested inheritance.
- get_config()[source]#
Get config flags for self.
- Returns:
- config_dictdict
Dictionary of config name : config value pairs. Collected from _config class attribute via nested inheritance and then any overrides and new tags from _onfig_dynamic object attribute.
- classmethod get_param_defaults()[source]#
Get object’s parameter defaults.
- Returns:
- default_dict: dict[str, Any]
Keys are all parameters of cls that have a default defined in __init__ values are the defaults, as defined in __init__.
- classmethod get_param_names()[source]#
Get object’s parameter names.
- Returns:
- param_names: list[str]
Alphabetically sorted list of parameter names of cls.
- get_params(deep=True)[source]#
Get a dict of parameters values for this object.
- Parameters:
- deepbool, default=True
Whether to return parameters of components.
If True, will return a dict of parameter name : value for this object, including parameters of components (= BaseObject-valued parameters).
If False, will return a dict of parameter name : value for this object, but not include parameters of components.
- Returns:
- paramsdict with str-valued keys
Dictionary of parameters, paramname : paramvalue keys-value pairs include:
always: all parameters of this object, as via get_param_names values are parameter value for that key, of this object values are always identical to values passed at construction
if deep=True, also contains keys/value pairs of component parameters parameters of components are indexed as [componentname]__[paramname] all parameters of componentname appear as paramname with its value
if deep=True, also contains arbitrary levels of component recursion, e.g., [componentname]__[componentcomponentname]__[paramname], etc
- get_tag(tag_name, tag_value_default=None, raise_error=True)[source]#
Get tag value from estimator class and dynamic tag overrides.
- Parameters:
- tag_namestr
Name of tag to be retrieved
- tag_value_defaultany type, optional; default=None
Default/fallback value if tag is not found
- raise_errorbool
whether a ValueError is raised when the tag is not found
- Returns:
- tag_valueAny
Value of the tag_name tag in self. If not found, returns an error if raise_error is True, otherwise it returns tag_value_default.
- Raises:
- ValueError if raise_error is True i.e. if tag_name is not in
- self.get_tags().keys()
- get_tags()[source]#
Get tags from estimator class and dynamic tag overrides.
- Returns:
- collected_tagsdict
Dictionary of tag name : tag value pairs. Collected from _tags class attribute via nested inheritance and then any overrides and new tags from _tags_dynamic object attribute.
- classmethod get_test_params(parameter_set='default')[source]#
Return testing parameter settings for the estimator.
- Parameters:
- parameter_setstr, default=”default”
Name of the set of test parameters to return, for use in tests. If no special parameters are defined for a value, will return “default” set.
- Returns:
- paramsdict or list of dict, default = {}
Parameters to create testing instances of the class Each dict are parameters to construct an “interesting” test instance, i.e., MyClass(**params) or MyClass(**params[i]) creates a valid test instance. create_test_instance uses the first (or only) dictionary in params
- is_composite()[source]#
Check if the object is composed of other BaseObjects.
A composite object is an object which contains objects, as parameters. Called on an instance, since this may differ by instance.
- Returns:
- composite: bool
Whether an object has any parameters whose values are BaseObjects.
- reset()[source]#
Reset the object to a clean post-init state.
Using reset, runs __init__ with current values of hyper-parameters (result of get_params). This Removes any object attributes, except:
hyper-parameters = arguments of __init__
object attributes containing double-underscores, i.e., the string “__”
Class and object methods, and class attributes are also unaffected.
- Returns:
- self
Instance of class reset to a clean post-init state but retaining the current hyper-parameter values.
Notes
Equivalent to sklearn.clone but overwrites self. After self.reset() call, self is equal in value to type(self)(**self.get_params(deep=False))
- set_config(**config_dict)[source]#
Set config flags to given values.
- Parameters:
- config_dictdict
Dictionary of config name : config value pairs.
- Returns:
- selfreference to self.
Notes
Changes object state, copies configs in config_dict to self._config_dynamic.
- set_params(**params)[source]#
Set the parameters of this object.
The method works on simple estimators as well as on composite objects. Parameter key strings
<component>__<parameter>
can be used for composites, i.e., objects that contain other objects, to access<parameter>
in the component<component>
. The string<parameter>
, without<component>__
, can also be used if this makes the reference unambiguous, e.g., there are no two parameters of components with the name<parameter>
.- Parameters:
- **paramsdict
BaseObject parameters, keys must be
<component>__<parameter>
strings. __ suffixes can alias full strings, if unique among get_params keys.
- Returns:
- selfreference to self (after parameters have been set)
- set_random_state(random_state=None, deep=True, self_policy='copy')[source]#
Set random_state pseudo-random seed parameters for self.
Finds
random_state
named parameters viaestimator.get_params
, and sets them to integers derived fromrandom_state
viaset_params
. These integers are sampled from chain hashing viasample_dependent_seed
, and guarantee pseudo-random independence of seeded random generators.Applies to
random_state
parameters inestimator
depending onself_policy
, and remaining component estimators if and only ifdeep=True
.Note: calls
set_params
even ifself
does not have arandom_state
, or none of the components have arandom_state
parameter. Therefore,set_random_state
will reset anyscikit-base
estimator, even those without arandom_state
parameter.- Parameters:
- random_stateint, RandomState instance or None, default=None
Pseudo-random number generator to control the generation of the random integers. Pass int for reproducible output across multiple function calls.
- deepbool, default=True
Whether to set the random state in sub-estimators. If False, will set only
self
’srandom_state
parameter, if exists. If True, will setrandom_state
parameters in sub-estimators as well.- self_policystr, one of {“copy”, “keep”, “new”}, default=”copy”
“copy” :
estimator.random_state
is set to inputrandom_state
“keep” :
estimator.random_state
is kept as is“new” :
estimator.random_state
is set to a new random state,
derived from input
random_state
, and in general different from it
- Returns:
- selfreference to self