Fix remove scikit learn

README.md

@@ -21,36 +21,33 @@
 
 <p align="center">Frouros is a Python library for drift detection in Machine Learning problems.</p>
 
-Frouros provides a combination of classical and more recent algorithms for drift detection, both for the supervised and unsupervised parts, as well as some semi-supervised algorithms. The library tries to fulfill two main objectives: 1. to be able to easily integrate in a machine learning model development pipeline with the [scikit-learn](https://github.com/scikit-learn/scikit-learn) library; 2. to unify in a single library the part of concept drift detection and
-adaptation (traditionally researched and used for streaming/evolving data streams and incremental learning) with the research of change detection in the covariate distributions (also known as data shift, related to the field of statistical two-sample testing and methods that measure distance between distributions).
+Frouros provides a combination of classical and more recent algorithms for drift detection, both for detecting concept and data drift.
 
 ## Quickstart
 
-As a quick and easy example, we can generate two bivariate normal distribution in order to use an unsupervised method like MMD (Maximum Mean Discrepancy). This method tries to verify if generated samples come from the same distribution or not. If they come from different distributions, it means that there is data drift.
+As a quick and easy example, we can generate two normal distributions in order to use a data drift detector like Kolmogorov-Smirnov. This method tries to verify if generated samples come from the same distribution or not. If they come from different distributions, it means that there is data drift.
 
 ```python
-from sklearn.gaussian_process.kernels import RBF
 import numpy as np
-from frouros.unsupervised.distance_based import MMD
+from frouros.data_drift.batch import KSTest
 
 np.random.seed(31)
-# X samples from a normal distribution with mean = [1. 1.] and cov = [[2. 0.][0. 2.]]
-x_mean = np.ones(2)
-x_cov = 2*np.eye(2)
-# Y samples a normal distribution with mean = [0. 0.] and cov = [[2. 1.][1. 2.]]
-y_mean = np.zeros(2)
-y_cov = np.eye(2) + 1
+# X samples from a normal distribution with mean=2 and std=2
+x_mean = 2
+x_std = 2
+# Y samples a normal distribution with mean=1 and std=2
+y_mean = 1
+y_std = 2
 
-num_samples = 200
-X_ref = np.random.multivariate_normal(x_mean, x_cov, num_samples)
-X_test = np.random.multivariate_normal(y_mean, y_cov, num_samples)
+num_samples = 10000
+X_ref = np.random.normal(x_mean, x_std, num_samples)
+X_test = np.random.normal(y_mean, y_std, num_samples)
 
 alpha = 0.01  # significance level for the hypothesis test
 
-detector = MMD(num_permutations=1000, kernel=RBF(length_scale=1.0), random_state=31)
+detector = KSTest()
 detector.fit(X=X_ref)
-detector.transform(X=X_test)
-mmd, p_value = detector.distance
+statistic, p_value = detector.compare(X=X_test)
 
 p_value < alpha
 >>> True  # Drift detected. We can reject H0, so both samples come from different distributions.
@@ -65,10 +62,7 @@ Frouros supports Python 3.8, 3.9 and 3.10 versions. It can be installed via pip:
 ```bash
 pip install frouros
 ```
-there is also the option to use [PyTorch](https://github.com/pytorch/pytorch) models with the help of [skorch](https://github.com/skorch-dev/skorch):
-```bash
-pip install frouros[pytorch]
-```
+
 Latest main branch modifications can be installed via:
 ```bash
 pip install git+https://github.com/IFCA/frouros.git
@@ -89,9 +83,9 @@ The currently supported methods are listed in the following table. They are divi
 <tbody>
   <tr>
     <td rowspan="12">
-        <a href="https://github.com/jaime-cespedes-sisniega/frouros/blob/main/frouros/supervised/base.py"> 
+        <a href="https://github.com/jaime-cespedes-sisniega/frouros/blob/main/frouros/concept_drift/base.py"> 
             <div style="height:100%;width:100%">
-                Supervised
+                Concept drift
             </div>
         </a>
     </td>
@@ -219,43 +213,11 @@ The currently supported methods are listed in the following table. They are divi
     </td>
   </tr>
   </tr>
-  <tr>
-    <td rowspan="2">
-        <a href="https://github.com/jaime-cespedes-sisniega/frouros/blob/main/frouros/semi_supervised/base.py"> 
-            <div style="height:100%;width:100%">
-                Semi-supervised
-            </div>
-        </a>
-    </td>
-    <td rowspan="2">
-        <a href="https://github.com/jaime-cespedes-sisniega/frouros/blob/main/frouros/semi_supervised/margin_density_based/base.py"> 
-            <div style="height:100%;width:100%">
-                Margin Density Based
-            </div>
-        </a>
-    </td>
-    <td>
-        <a href="https://github.com/jaime-cespedes-sisniega/frouros/blob/main/frouros/semi_supervised/margin_density_based/md3.py"> 
-            <div style="height:100%;width:100%">
-                MD3-SVM
-            </div>
-        </a>
-    </td>
-  <tr>
-    <td>
-        <a href="https://github.com/jaime-cespedes-sisniega/frouros/blob/main/frouros/supervised/margin_density_based/md3.py">  
-            <div style="height:100%;width:100%">
-                MD3-RS
-            </div>
-        </a>
-    </td>
-  </tr>
-  </tr>
   <tr>
     <td rowspan="10">
-        <a href="https://github.com/jaime-cespedes-sisniega/frouros/blob/main/frouros/unsupervised/base.py"> 
+        <a href="https://github.com/jaime-cespedes-sisniega/frouros/blob/main/frouros/data_drift/base.py"> 
             <div style="height:100%;width:100%">
-                Unsupervised
+                Data drift
             </div>
         </a>
     </td>

frouros/concept_drift/__init__.py

@@ -0,0 +1,59 @@
+"""Concept drift detection methods init."""
+
+from .cusum_based import (
+    CUSUM,
+    CUSUMConfig,
+    GeometricMovingAverage,
+    GeometricMovingAverageConfig,
+    PageHinkley,
+    PageHinkleyConfig,
+)
+from .ddm_based import (
+    DDM,
+    DDMConfig,
+    ECDDWT,
+    ECDDWTConfig,
+    EDDM,
+    EDDMConfig,
+    HDDMA,
+    HDDMAConfig,
+    HDDMW,
+    HDDMWConfig,
+    RDDM,
+    RDDMConfig,
+    STEPD,
+    STEPDConfig,
+)
+from .window_based import (
+    ADWIN,
+    ADWINConfig,
+    KSWIN,
+    KSWINConfig,
+)
+
+__all__ = [
+    "ADWIN",
+    "ADWINConfig",
+    "CUSUM",
+    "CUSUMConfig",
+    "DDM",
+    "DDMConfig",
+    "ECDDWT",
+    "ECDDWTConfig",
+    "EDDM",
+    "EDDMConfig",
+    "GeometricMovingAverage",
+    "GeometricMovingAverageConfig",
+    "HDDMA",
+    "HDDMAConfig",
+    "HDDMW",
+    "HDDMWConfig",
+    "KSWIN",
+    "KSWINConfig",
+    "PageHinkley",
+    "PageHinkleyConfig",
+    "RDDM",
+    "RDDMConfig",
+    "STEPD",
+    "STEPDConfig",
+]

frouros/concept_drift/base.py

@@ -0,0 +1,118 @@
+"""Concept drift base module."""
+
+import abc
+from typing import (  # noqa: TYP001
+    Dict,
+    Union,
+)
+
+
+class ConceptDriftBaseConfig(abc.ABC):
+    """Abstract class representing a concept drift configuration class."""
+
+    def __init__(
+        self,
+        min_num_instances: int,
+    ) -> None:
+        """Init method.
+
+        :param min_num_instances: minimum numbers of instances
+        to start looking for changes
+        :type min_num_instances: int
+        """
+        self.min_num_instances = min_num_instances
+
+    @property
+    def min_num_instances(self) -> int:
+        """Minimum number of instances property.
+
+        :return: minimum number of instances to start looking for changes
+        :rtype: int
+        """
+        return self._min_num_instances
+
+    @min_num_instances.setter
+    def min_num_instances(self, value: int) -> None:
+        """Minimum number of instances setter.
+
+        :param value: value to be set
+        :type value: int
+        """
+        self._min_num_instances = value
+
+
+class ConceptDriftBase(abc.ABC):
+    """Abstract class representing a delayed target."""
+
+    def __init__(
+        self,
+        config: ConceptDriftBaseConfig,
+    ) -> None:
+        """Init method.
+
+        :param config: configuration parameters
+        :type config: ConceptDriftBaseConfig
+        """
+        self.config = config
+        self.num_instances = 0
+
+    @property
+    def config(self) -> ConceptDriftBaseConfig:
+        """Config property.
+
+        :return: configuration parameters of the estimator
+        :rtype: ConceptDriftBaseConfig
+        """
+        return self._config
+
+    @config.setter
+    def config(self, value: ConceptDriftBaseConfig) -> None:
+        """Config setter.
+
+        :param value: value to be set
+        :type value: ConceptDriftBaseConfig
+        :raises TypeError: Type error exception
+        """
+        if not isinstance(value, ConceptDriftBaseConfig):
+            raise TypeError("value must be of type ConceptDriftBaseConfig.")
+        self._config = value
+
+    @property
+    def num_instances(self) -> int:
+        """Number of instances counter property.
+
+        :return: Number of instances counter value
+        :rtype: int
+        """
+        return self._num_instances
+
+    @num_instances.setter
+    def num_instances(self, value: int) -> None:
+        """Number of instances counter setter.
+
+        :param value: value to be set
+        :type value: int
+        :raises ValueError: Value error exception
+        """
+        if value < 0:
+            raise ValueError("num_instances must be greater or equal than 0.")
+        self._num_instances = value
+
+    def reset(self, *args, **kwargs) -> None:
+        """Reset method."""
+
+    @property
+    def status(self) -> Dict[str, bool]:
+        """Status property.
+
+        :return: status dict
+        :rtype: Dict[str, bool]
+        """
+
+    @abc.abstractmethod
+    def update(self, value: Union[int, float]) -> None:
+        """Abstract update method.
+
+        :param value: value to update detector
+        :type value: Union[int, float]
+        """

frouros/supervised/cusum_based/__init__.py → frouros/concept_drift/cusum_based/__init__.py

@@ -1,4 +1,4 @@
-"""Supervised CUSUM based detection methods' init."""
+"""Concept drift CUSUM based detection methods' init."""
 
 from .cusum import CUSUM, CUSUMConfig
 from .geometric_moving_average import (

frouros/supervised/cusum_based/base.py → frouros/concept_drift/cusum_based/base.py

@@ -1,18 +1,16 @@
-"""Supervised CUSUM based base module."""
+"""Concept drift CUSUM based base module."""
 
 import abc
 from typing import (  # noqa: TYP001
     Dict,
     Union,
 )
 
-from sklearn.base import BaseEstimator  # type: ignore
-
-from frouros.supervised.base import SupervisedBaseEstimator, SupervisedBaseConfig
+from frouros.concept_drift.base import ConceptDriftBase, ConceptDriftBaseConfig
 from frouros.utils.stats import Mean
 
 
-class CUSUMBaseConfig(SupervisedBaseConfig):
+class CUSUMBaseConfig(ConceptDriftBaseConfig):
     """Class representing a CUSUM based configuration class."""
 
     def __init__(
@@ -125,22 +123,19 @@ def alpha(self, value: float) -> None:
         self._alpha = value
 
 
-class CUSUMBaseEstimator(SupervisedBaseEstimator):
+class CUSUMBase(ConceptDriftBase):
     """CUSUM based algorithm class."""
 
     def __init__(
         self,
-        estimator: BaseEstimator,
         config: CUSUMBaseConfig,
     ) -> None:
         """Init method.
 
-        :param estimator: sklearn estimator
-        :type estimator: BaseEstimator
         :param config: configuration parameters
         :type config: CUSUMBaseConfig
         """
-        super().__init__(estimator=estimator, config=config)
+        super().__init__(config=config)
         self.mean_error_rate = Mean()
         self.sum_ = 0.0
         self.drift = False