Migrate to v1.1.4

README.md

@@ -32,15 +32,15 @@ import numpy as np
 from sklearn.datasets import make_classification
 from sklearn.model_selection import train_test_split
 from sklearn.preprocessing import StandardScaler
-from hwm.estimators import HammersteinWienerClassifier
+from hwm.estimators import HWClassifier
 from hwm.metrics import prediction_stability_score
 
 # Generate synthetic data
 X, y = make_classification(n_samples=10000, n_features=20)
 X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
 
 # Initialize the model
-hw_model = HammersteinWienerClassifier(
+hw_model = HWClassifier(
     nonlinear_input_estimator=StandardScaler(),
     nonlinear_output_estimator=StandardScaler(),
     p=2,

doc/source/estimators.rst

@@ -146,7 +146,7 @@ for fine-tuning the model's behavior during training. These include:
 Hammerstein-Wiener Classifier
 ===============================
 
-The `HammersteinWienerClassifier` is a dynamic system model for 
+The `HammersteinWienerClassifier`( :class:`hwm.estimators.HWClassifier` ) is a dynamic system model for 
 classification tasks. It utilizes the Hammerstein-Wiener model 
 structure, which is composed of a nonlinear input block, a linear 
 dynamic block, and a nonlinear output block. This allows it to capture 
@@ -225,7 +225,7 @@ and a system dynamics dataset.
     from sklearn.preprocessing import StandardScaler
     from sklearn.datasets import make_classification
     from sklearn.model_selection import train_test_split
-    from hwm.estimators import HammersteinWienerClassifier
+    from hwm.estimators import HWClassifier
 
     # Generate synthetic data
     X, y = make_classification(n_samples=1000, n_features=20, 
@@ -235,7 +235,7 @@ and a system dynamics dataset.
     )
 
     # Initialize classifier with nonlinear transformations
-    hw_classifier = HammersteinWienerClassifier(
+    hw_classifier = HWClassifier(
         nonlinear_input_estimator=StandardScaler(),
         nonlinear_output_estimator=StandardScaler(),
         p=2,
@@ -277,7 +277,7 @@ performance using `twa_score` and `prediction_stability_score`.
     import numpy as np
     from sklearn.preprocessing import StandardScaler
     from sklearn.model_selection import train_test_split
-    from hwm.estimators import HammersteinWienerClassifier
+    from hwm.estimators import HWClassifier
     from hwm.datasets import make_system_dynamics
     from hwm.metrics import twa_score, prediction_stability_score
     
@@ -303,7 +303,7 @@ performance using `twa_score` and `prediction_stability_score`.
     )
 
     # Initialize classifier
-    hw_classifier = HammersteinWienerClassifier(
+    hw_classifier = HWClassifier(
         p=3,
         nonlinear_input_estimator=StandardScaler(),
         nonlinear_output_estimator=StandardScaler(),
@@ -378,7 +378,7 @@ performance using `twa_score` and `prediction_stability_score`.
 Hammerstein-Wiener Regressor
 ==============================
 
-The :class:`~hwm.estimators.dynamic_system.HammersteinWienerRegressor` class implements a nonlinear regression 
+The :class:`~hwm.estimators.dynamic_system.HWRegressor` class implements a nonlinear regression 
 model based on the Hammerstein-Wiener (HW) architecture. This block-structured model combines a nonlinear 
 input transformation, a linear dynamic system block, and a nonlinear output transformation, making it highly 
 suitable for regression tasks where data exhibit complex, time-dependent relationships. The HW model is 
@@ -443,19 +443,19 @@ Attributes
 Example Usage
 ---------------
 
-Below are example applications of `HammersteinWienerRegressor`, 
+Below are example applications of :class:`hwm.estimators.HWRegressor`, 
 demonstrating initialization, training, and evaluation on synthetic data. 
 The first example shows time-series data, while the second highlights 
 financial trend forecasting.
 
 **Example 1: Basic Time-Series Regression**
 
-This example demonstrates using `HammersteinWienerRegressor` to train 
+This example demonstrates using `HWRegressor` to train 
 and predict on synthetic time-series data.
 
 .. code-block:: python
 
-    from hwm.estimators import HammersteinWienerRegressor
+    from hwm.estimators import HWRegressor
     from sklearn.preprocessing import StandardScaler
     from sklearn.datasets import make_regression
     from sklearn.model_selection import train_test_split
@@ -498,12 +498,12 @@ and predict on synthetic time-series data.
 
 **Example 2: Financial Market Trend Forecasting**
 
-This example demonstrates using `HammersteinWienerRegressor` on synthetic financial 
+This example demonstrates using `HWRegressor` on synthetic financial 
 trend data, emphasizing the model’s adaptability to financial forecasting tasks.
 
 .. code-block:: python
 
-    from hwm.estimators import HammersteinWienerRegressor
+    from hwm.estimators import HWRegressor
     from sklearn.preprocessing import StandardScaler
     from sklearn.model_selection import train_test_split
     from hwm.datasets import make_financial_market_trends
@@ -530,7 +530,7 @@ trend data, emphasizing the model’s adaptability to financial forecasting task
     X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
 
     # Initialize Hammerstein-Wiener Regressor
-    hw_regressor = HammersteinWienerRegressor(
+    hw_regressor = HWRegressor(
         p=3,
         nonlinear_input_estimator=StandardScaler(),
         nonlinear_output_estimator=StandardScaler(),
@@ -580,7 +580,7 @@ trend data, emphasizing the model’s adaptability to financial forecasting task
 
 .. seealso::
 
-    - :class:`~hwm.estimators.HammersteinWienerClassifier`
+    - :class:`~hwm.estimators.HWClassifier`
     - :class:`~sklearn.linear_model.SGDRegressor`
 
 

doc/source/release_notes.rst

@@ -1,10 +1,10 @@
 .. _release_notes:
 
 Release Notes
-=============
+===============
 
-hwm v1.1.0
-----------
+hwm v1.1.4
+------------
 
 **Released on 2024-04-27**
 
@@ -51,26 +51,77 @@ hwm v1.1.0
   .. code-block:: python
 
       import numpy as np
-      from hwm.estimators import HammersteinWienerRegressor
+      from hwm.estimators import HWRegressor
 
       # Generate large synthetic dataset
       X = np.random.rand(1000000, 10)
       y = np.random.rand(1000000)
 
       # Initialize and fit the regressor with batch computation
-      model = HammersteinWienerRegressor(batch_size=10000)
+      model = HWRegressor(batch_size=10000)
       model.fit(X, y)
 
       # Make predictions
       predictions = model.predict(X)
       print(predictions[:5])
       # Output: [0.523, 0.489, 0.501, 0.478, 0.495]
 
+
+
+**Enhancements**
+
+- *Renamed Classes for Improved Usability*
+
+  The class names :class:`hwm.estimators.HammersteinWienerRegressor` and 
+  :class:`hwm.estimators.HammersteinWienerClassifier` have been renamed to 
+  :class:`hwm.estimators.HWRegressor` and :class:`hwm.estimators.HWClassifier` 
+  respectively for brevity and ease of use.
+
+  **Deprecation Notice:**
+
+  - The old class names `HammersteinWienerRegressor` and 
+    `HammersteinWienerClassifier` are now deprecated and will 
+    be removed in version **1.1.3**. Users are encouraged to 
+    transition to the new class names to ensure future compatibility.
+
+  *Example:*
+
+  .. code-block:: python
+
+      from hwm.estimators import HWRegressor, HWClassifier
+
+      # Initialize the regressor
+      regressor = HWRegressor(batch_size=10000)
+      regressor.fit(X_train, y_train)
+      predictions = regressor.predict(X_test)
+
+      # Initialize the classifier
+      classifier = HWClassifier(batch_size=10000)
+      classifier.fit(X_train, y_train)
+      class_predictions = classifier.predict(X_test)
+      class_probabilities = classifier.predict_proba(X_test)
+
+  **Using Deprecated Class Names:**
+
+  .. code-block:: python
+
+      import warnings
+      from hwm.estimators import HammersteinWienerRegressor
+
+      # To display deprecation warnings
+      warnings.simplefilter('default', DeprecationWarning)
+
+      # Initialize the deprecated regressor
+      regressor = HammersteinWienerRegressor(batch_size=5000)
+      # Output: DeprecationWarning: HammersteinWienerRegressor is deprecated 
+      # and will be removed in version 1.2. Use HWRegressor instead.
+      
+      
 **Bug Fixes**
 
-- **Resolved Memory Errors in :class:`hwm.estimators.HammersteinWienerRegressor`**
+- **Resolved Memory Errors in :class:`hwm.estimators.HWRegressor` **
 
-  Addressed memory consumption issues in the `HammersteinWienerRegressor` when processing 
+  Addressed memory consumption issues in the `HWRegressor` when processing 
   large datasets by introducing batch processing mechanisms. This fix ensures stable and 
   efficient model training and prediction without exhausting system memory.
 
@@ -83,8 +134,8 @@ hwm v1.1.0
 
 **Upgrade Notes**
 
-- Users upgrading from version 1.0.1 to 1.1.0 should ensure that their workflows 
-  accommodate the new batch processing parameters in :class:`hwm.estimators.HammersteinWienerRegressor`.
+- Users upgrading from version 1.0.1 to 1.1.4 should ensure that their workflows 
+  accommodate the new batch processing parameters in :class:`hwm.estimators.HWRegressor`.
 - The :func:`hwm.metrics.twa_score` function now accepts both label arrays and 
   probability arrays. Ensure that the input 
   format aligns with the desired usage.

examples/kdd_cup_1999.py

@@ -52,7 +52,7 @@
 from tensorflow.keras.layers import Dense, LSTM
 from tensorflow.keras.models import Sequential
 
-from hwm.estimators import HammersteinWienerClassifier
+from hwm.estimators import HWClassifier
 from hwm.metrics import prediction_stability_score, twa_score
 from hwm.utils import resample_data
 
@@ -145,7 +145,7 @@ def transform(self, X):
 
 # %%
 # Initialize the Hammerstein-Wiener Classifier
-hw_model = HammersteinWienerClassifier(
+hw_model = HWClassifier(
     nonlinear_input_estimator=ReLUTransformer(),
     nonlinear_output_estimator=ReLUTransformer(),
     p=9,
@@ -176,7 +176,7 @@ def transform(self, X):
 }
 
 # Initialize the Hammerstein-Wiener Classifier with fixed components
-fixed_hw_model = HammersteinWienerClassifier(
+fixed_hw_model = HWClassifier(
     nonlinear_input_estimator=ReLUTransformer(),
     nonlinear_output_estimator=ReLUTransformer(),
     loss="cross_entropy",

hwm/__init__.py

@@ -38,7 +38,7 @@ def _lazy_loader():
 try:
     from ._version import version as __version__
 except ImportError:
-    __version__ = "1.1.0"
+    __version__ = "1.1.4"
 
 # Dependency management
 _required_dependencies = [
@@ -66,6 +66,7 @@ def _lazy_loader():
 
 # Suppression of non-critical warnings, adjustable by the user
 _warnings_state = {"FutureWarning": "ignore"}
+
 def suppress_warnings(suppress=True):
     """Enable or disable future/syntax warnings."""
     for warning, action in _warnings_state.items():