Add concatenation support for multiple feature libraries

pysindy/feature_library/__init__.py

@@ -1,3 +1,4 @@
+from .concat_library import ConcatLibrary
 from .custom_library import CustomLibrary
 from .fourier_library import FourierLibrary
 from .identity_library import IdentityLibrary

pysindy/feature_library/concat_library.py

@@ -0,0 +1,121 @@
+import numpy as np
+
+from .feature_library import BaseFeatureLibrary
+
+
+class ConcatLibrary(BaseFeatureLibrary):
+    """Concatenate multiple libraries into one library. All settings
+    provided to individual libraries will be applied.
+
+    Parameters
+    ----------
+    libraries : list of libraries
+        Library instances to be applied to the input matrix.
+
+    Attributes
+    ----------
+    libraries_ : list of libraries
+        Library instances to be applied to the input matrix.
+
+    n_output_features_ : int
+        The total number of output features. The number of output features
+        is the product of the number of library functions and the number of
+        input features.
+
+    Examples
+    --------
+    >>> import numpy as np
+    >>> from pysindy.feature_library import FourierLibrary, CustomLibrary
+    >>> from pysindy.feature_library import ConcatLibrary
+    >>> X = np.array([[0.,-1],[1.,0.],[2.,-1.]])
+    >>> functions = [lambda x : np.exp(x), lambda x,y : np.sin(x+y)]
+    >>> lib_custom = CustomLibrary(library_functions=functions)
+    >>> lib_fourier = FourierLibrary()
+    >>> lib_concat = ConcatLibrary([lib_custom, lib_fourier])
+    >>> lib_concat.fit()
+    >>> lib.transform(X)
+    """
+
+    def __init__(self, libraries: list):
+        super(ConcatLibrary, self).__init__()
+        self.libraries_ = libraries
+
+    def fit(self, X, y=None):
+        """
+        Compute number of output features.
+
+        Parameters
+        ----------
+        X : array-like, shape (n_samples, n_features)
+            The data.
+
+        Returns
+        -------
+        self : instance
+        """
+
+        # first fit all libs provided below
+        fitted_libs = [lib.fit(X, y) for lib in self.libraries_]
+
+        # calculate the sum of output features
+        self.n_output_features_ = sum([lib.n_output_features_ for lib in fitted_libs])
+
+        # save fitted libs
+        self.libraries_ = fitted_libs
+
+        return self
+
+    def transform(self, X):
+        """Transform data with libs provided below.
+
+        Parameters
+        ----------
+        X : array-like, shape [n_samples, n_features]
+            The data to transform, row by row.
+
+        Returns
+        -------
+        XP : np.ndarray, shape [n_samples, NP]
+            The matrix of features, where NP is the number of features
+            generated from applying the custom functions to the inputs.
+
+        """
+
+        n_samples = X.shape[0]
+
+        # preallocate matrix
+        XP = np.zeros((n_samples, self.n_output_features_))
+
+        current_feat = 0
+        for lib in self.libraries_:
+
+            # retrieve num features from lib
+            lib_n_output_features = lib.n_output_features_
+
+            start_feature_index = current_feat
+            end_feature_index = start_feature_index + lib_n_output_features
+
+            XP[:, start_feature_index:end_feature_index] = lib.transform(X)
+
+            current_feat += lib_n_output_features
+
+        return XP
+
+    def get_feature_names(self, input_features=None):
+        """Return feature names for output features.
+
+        Parameters
+        ----------
+        input_features : list of string, length n_features, optional
+            String names for input features if available. By default,
+            "x0", "x1", ... "xn_features" is used.
+
+        Returns
+        -------
+        output_feature_names : list of string, length n_output_features
+        """
+        feature_names = list()
+        for lib in self.libraries_:
+            lib_feat_names = lib.get_feature_names(input_features)
+            feature_names += lib_feat_names
+        return feature_names

test/feature_library/test_feature_library.py

@@ -8,6 +8,7 @@
 from sklearn.exceptions import NotFittedError
 from sklearn.utils.validation import check_is_fitted
 
+from pysindy.feature_library import ConcatLibrary
 from pysindy.feature_library import CustomLibrary
 from pysindy.feature_library import FourierLibrary
 from pysindy.feature_library import IdentityLibrary
@@ -59,6 +60,7 @@ def test_bad_parameters():
         IdentityLibrary(),
         PolynomialLibrary(),
         FourierLibrary(),
+        ConcatLibrary([IdentityLibrary(), PolynomialLibrary()]),
         pytest.lazy_fixture("data_custom_library"),
     ],
 )
@@ -74,6 +76,7 @@ def test_fit_transform(data_lorenz, library):
         IdentityLibrary(),
         PolynomialLibrary(),
         FourierLibrary(),
+        ConcatLibrary([IdentityLibrary(), PolynomialLibrary()]),
         pytest.lazy_fixture("data_custom_library"),
     ],
 )
@@ -89,6 +92,7 @@ def test_change_in_data_shape(data_lorenz, library):
     [
         (IdentityLibrary(), 3),
         (PolynomialLibrary(), 10),
+        (ConcatLibrary([IdentityLibrary(), PolynomialLibrary()]), 13),
         (FourierLibrary(), 6),
         (pytest.lazy_fixture("data_custom_library"), 9),
     ],
@@ -107,6 +111,7 @@ def test_output_shape(data_lorenz, library, shape):
         IdentityLibrary(),
         PolynomialLibrary(),
         FourierLibrary(),
+        ConcatLibrary([PolynomialLibrary(), FourierLibrary()]),
         pytest.lazy_fixture("data_custom_library"),
     ],
 )
@@ -174,3 +179,10 @@ def test_not_implemented(data_lorenz):
 
     with pytest.raises(NotImplementedError):
         library.get_feature_names(x)
+
+
+def test_concat():
+    ident_lib = IdentityLibrary()
+    poly_lib = PolynomialLibrary()
+    concat_lib = ConcatLibrary([ident_lib, poly_lib])
+    assert isinstance(concat_lib, ConcatLibrary)