fix: standardize variable names to X and Y for consistency (#206)

xeofs/models/_base_model.py

@@ -186,6 +186,6 @@ def load(
         model = cls.deserialize(dt)
         return model
 
-    def _validate_loaded_data(self, data: DataArray):
+    def _validate_loaded_data(self, X: DataArray):
         """Optionally check the loaded data for placeholders."""
         pass

xeofs/models/_base_model_cross_set.py

@@ -343,11 +343,11 @@ def transform(
         data = self._transform_algorithm(X, Y, normalized=normalized)
         data_list = []
         if X is not None:
-            X = self.whitener1.inverse_transform_scores_unseen(data["data1"])
+            X = self.whitener1.inverse_transform_scores_unseen(data["X"])
             X = self.preprocessor1.inverse_transform_scores_unseen(X)
             data_list.append(X)
         if Y is not None:
-            Y = self.whitener2.inverse_transform_scores_unseen(data["data2"])
+            Y = self.whitener2.inverse_transform_scores_unseen(data["Y"])
             Y = self.preprocessor2.inverse_transform_scores_unseen(Y)
             data_list.append(Y)
 

xeofs/models/cpcca.py

@@ -237,7 +237,7 @@ def _transform_algorithm(
             scores1 = xr.dot(X, comps1)
             if normalized:
                 scores1 = scores1 / norm1
-            results["data1"] = scores1
+            results["X"] = scores1
 
         if Y is not None:
             # Project data onto singular vectors
@@ -246,7 +246,7 @@ def _transform_algorithm(
             scores2 = xr.dot(Y, comps2)
             if normalized:
                 scores2 = scores2 / norm2
-            results["data2"] = scores2
+            results["Y"] = scores2
 
         return results
 

xeofs/models/eof.py

@@ -47,7 +47,7 @@ class EOF(_BaseModelSingleSet):
     Examples
     --------
     >>> model = xe.models.EOF(n_modes=5)
-    >>> model.fit(data)
+    >>> model.fit(X)
     >>> scores = model.scores()
 
     """
@@ -83,32 +83,32 @@ def __init__(
         )
         self.attrs.update({"model": "EOF analysis"})
 
-    def _fit_algorithm(self, data: DataArray) -> Self:
+    def _fit_algorithm(self, X: DataArray) -> Self:
         sample_name = self.sample_name
         feature_name = self.feature_name
 
         # Augment the data
-        data = self._augment_data(data)
+        X = self._augment_data(X)
 
         # Compute the total variance
-        total_variance = compute_total_variance(data, dim=sample_name)
+        total_variance = compute_total_variance(X, dim=sample_name)
 
         # Decompose the data
         decomposer = Decomposer(**self._decomposer_kwargs)
-        decomposer.fit(data, dims=(sample_name, feature_name))
+        decomposer.fit(X, dims=(sample_name, feature_name))
 
         singular_values = decomposer.s_
         components = decomposer.V_
         scores = decomposer.U_ * decomposer.s_
         scores.name = "scores"
 
         # Compute the explained variance per mode
-        n_samples = data.coords[self.sample_name].size
+        n_samples = X.coords[self.sample_name].size
         exp_var = singular_values**2 / (n_samples - 1)
         exp_var.name = "explained_variance"
 
         # Store the results
-        self.data.add(data, "input_data", allow_compute=False)
+        self.data.add(X, "input_data", allow_compute=False)
         self.data.add(components, "components")
         self.data.add(scores, "scores")
         self.data.add(singular_values, "norms")
@@ -118,16 +118,16 @@ def _fit_algorithm(self, data: DataArray) -> Self:
         self.data.set_attrs(self.attrs)
         return self
 
-    def _augment_data(self, data: DataArray) -> DataArray:
-        return data
+    def _augment_data(self, X: DataArray) -> DataArray:
+        return X
 
-    def _transform_algorithm(self, data: DataObject) -> DataArray:
+    def _transform_algorithm(self, X: DataObject) -> DataArray:
         feature_name = self.preprocessor.feature_name
 
         components = self.data["components"]
 
         # Project the data
-        projections = xr.dot(data, components, dims=feature_name)
+        projections = xr.dot(X, components, dims=feature_name)
         projections.name = "scores"
 
         return projections
@@ -333,13 +333,13 @@ def __init__(
         )
         self.attrs.update({"model": "Complex EOF analysis"})
 
-    def _fit_algorithm(self, data: DataArray) -> Self:
-        if not np.iscomplexobj(data):
+    def _fit_algorithm(self, X: DataArray) -> Self:
+        if not np.iscomplexobj(X):
             warnings.warn(
                 "Expected complex-valued data but found real-valued data. For Hilbert EOF analysis, use `HilbertEOF` model."
             )
 
-        return super()._fit_algorithm(data)
+        return super()._fit_algorithm(X)
 
     def components_amplitude(self) -> DataObject:
         """Return the amplitude of the (EOF) components.
@@ -496,7 +496,7 @@ class HilbertEOF(ComplexEOF):
     Examples
     --------
     >>> model = HilbertEOF(n_modes=5, standardize=True)
-    >>> model.fit(data)
+    >>> model.fit(X)
 
     """
 
@@ -534,22 +534,22 @@ def __init__(
         self.attrs.update({"model": "Hilbert EOF analysis"})
         self._params.update({"padding": padding, "decay_factor": decay_factor})
 
-    def _augment_data(self, data: DataArray) -> DataArray:
+    def _augment_data(self, X: DataArray) -> DataArray:
         # Apply hilbert transform:
         padding = self._params["padding"]
         decay_factor = self._params["decay_factor"]
         return hilbert_transform(
-            data,
+            X,
             dims=(self.sample_name, self.feature_name),
             padding=padding,
             decay_factor=decay_factor,
         )
 
-    def _fit_algorithm(self, data: DataArray) -> Self:
-        EOF._fit_algorithm(self, data)
+    def _fit_algorithm(self, X: DataArray) -> Self:
+        EOF._fit_algorithm(self, X)
         return self
 
-    def _transform_algorithm(self, data: DataArray) -> DataArray:
+    def _transform_algorithm(self, X: DataArray) -> DataArray:
         raise NotImplementedError("Hilbert EOF does not support transform method.")
 
     def _inverse_transform_algorithm(self, scores: DataArray) -> DataArray:

xeofs/models/eof_rotator.py

@@ -48,7 +48,7 @@ class EOFRotator(EOF):
     Examples
     --------
     >>> model = xe.models.EOF(n_modes=10)
-    >>> model.fit(data)
+    >>> model.fit(X, "time")
     >>> rotator = xe.models.EOFRotator(n_modes=10)
     >>> rotator.fit(model)
     >>> rotator.components()
@@ -222,7 +222,7 @@ def _sort_by_variance(self):
                     )
         self.sorted = True
 
-    def _transform_algorithm(self, data: DataArray) -> DataArray:
+    def _transform_algorithm(self, X: DataArray) -> DataArray:
         n_modes = self._params["n_modes"]
 
         svals = self.model.singular_values().sel(mode=slice(1, self._params["n_modes"]))
@@ -231,7 +231,7 @@ def _transform_algorithm(self, data: DataArray) -> DataArray:
         components = self.model.data["components"].sel(mode=slice(1, n_modes))
 
         # Compute non-rotated scores by projecting the data onto non-rotated components
-        projections = xr.dot(data, components) / svals
+        projections = xr.dot(X, components) / svals
         projections.name = "scores"
 
         # Rotate the scores

xeofs/models/opa.py

@@ -55,7 +55,7 @@ class OPA(_BaseModelSingleSet):
     --------
     >>> from xeofs.models import OPA
     >>> model = OPA(n_modes=10, tau_max=50, n_pca_modes=100)
-    >>> model.fit(data, dim=("time"))
+    >>> model.fit(X, dim=("time"))
 
     Retrieve the optimally persistent patterns (OPP) and their time series:
 
@@ -127,8 +127,8 @@ def _compute_matrix_inverse(X, dims):
             dask="allowed",
         )
 
-    def _fit_algorithm(self, data: DataArray) -> Self:
-        assert_not_complex(data)
+    def _fit_algorithm(self, X: DataArray) -> Self:
+        assert_not_complex(X)
 
         sample_name = self.sample_name
         feature_name = self.feature_name
@@ -146,8 +146,8 @@ def _fit_algorithm(self, data: DataArray) -> Self:
             check_nans=False,
             solver_kwargs=self._params["solver_kwargs"],
         )
-        pca.fit(data, dim=sample_name)
-        n_samples = data.coords[sample_name].size
+        pca.fit(X, dim=sample_name)
+        n_samples = X.coords[sample_name].size
         comps = pca.data["components"] * np.sqrt(n_samples - 1)
         # -> comps (feature x mode)
         scores = pca.data["scores"] / np.sqrt(n_samples - 1)
@@ -270,7 +270,7 @@ def _fit_algorithm(self, data: DataArray) -> Self:
         self._C0 = C0  # store C0 for testing purposes of orthogonality
         return self
 
-    def _transform_algorithm(self, data: DataArray) -> DataArray:
+    def _transform_algorithm(self, X: DataArray) -> DataArray:
         raise NotImplementedError("OPA does not (yet) support transform()")
 
     def _inverse_transform_algorithm(self, scores) -> DataObject:

xeofs/models/sparse_pca.py

@@ -143,24 +143,24 @@ def __init__(
             }
         )
 
-    def _fit_algorithm(self, data: DataArray) -> Self:
+    def _fit_algorithm(self, X: DataArray) -> Self:
         sample_name = self.sample_name
         feature_name = self.feature_name
 
         # Check if the data is real
         # NOTE: Complex data is not supported, it's likely possible but current numpy implementation
         # of sparse_pca needs to be adpated, mainly changing matrix transpose to conjugate transpose.
         # http://arxiv.org/abs/1804.00341
-        assert_not_complex(data)
+        assert_not_complex(X)
 
         # Compute the total variance
-        total_variance = compute_total_variance(data, dim=sample_name)
+        total_variance = compute_total_variance(X, dim=sample_name)
 
         # Compute matrix rank
-        rank = get_matrix_rank(data)
+        rank = get_matrix_rank(X)
 
         # Decide whether to use exact or randomized algorithm
-        is_small_data = max(data.shape) < 500
+        is_small_data = max(X.shape) < 500
         solver = self._params["solver"]
 
         match solver:
@@ -209,7 +209,7 @@ def _fit_algorithm(self, data: DataArray) -> Self:
         # exp_var : eigenvalues
         components, components_normal, exp_var = xr.apply_ufunc(
             decomposing_algorithm,
-            data,
+            X,
             input_core_dims=[[sample_name, feature_name]],
             output_core_dims=[[feature_name, "mode"], [feature_name, "mode"], ["mode"]],
             dask="allowed",
@@ -223,21 +223,21 @@ def _fit_algorithm(self, data: DataArray) -> Self:
         components.name = "sparse_weight_vectors"
         components = components.assign_coords(
             {
-                feature_name: data.coords[feature_name],
+                feature_name: X.coords[feature_name],
                 "mode": np.arange(1, self.n_modes + 1),
             },
         )
 
         components_normal.name = "orthonormal_weight_vectors"
         components_normal = components_normal.assign_coords(
             {
-                feature_name: data.coords[feature_name],
+                feature_name: X.coords[feature_name],
                 "mode": np.arange(1, self.n_modes + 1),
             },
         )
 
         # Transform the data
-        scores = xr.dot(data, components, dims=feature_name)
+        scores = xr.dot(X, components, dims=feature_name)
         scores.name = "scores"
 
         norms = xr.apply_ufunc(
@@ -253,7 +253,7 @@ def _fit_algorithm(self, data: DataArray) -> Self:
         norms.name = "component_norms"
 
         # Store the results
-        self.data.add(data, "input_data", allow_compute=False)
+        self.data.add(X, "input_data", allow_compute=False)
         self.data.add(components, "components")
         self.data.add(components_normal, "components_normal")
         self.data.add(scores, "scores")
@@ -264,13 +264,13 @@ def _fit_algorithm(self, data: DataArray) -> Self:
         self.data.set_attrs(self.attrs)
         return self
 
-    def _transform_algorithm(self, data: DataObject) -> DataArray:
+    def _transform_algorithm(self, X: DataObject) -> DataArray:
         feature_name = self.preprocessor.feature_name
 
         components = self.data["components"]
 
         # Project the data
-        projections = xr.dot(data, components, dims=feature_name)
+        projections = xr.dot(X, components, dims=feature_name)
         projections.name = "scores"
 
         return projections