chemotools 0.1.6__tar.gz → 0.1.7__tar.gz

{chemotools-0.1.6 → chemotools-0.1.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: chemotools
-Version: 0.1.6
+Version: 0.1.7
 Summary: chemotools: A Python Package that Integrates Chemometrics and scikit-learn
 License: MIT
 Author: Pau Cabaneros

{chemotools-0.1.6 → chemotools-0.1.7}/chemotools/augmentation/index_shift.py

@@ -1,6 +1,7 @@
-from typing import Optional
+from typing import Literal, Optional
 
 import numpy as np
+from numpy.polynomial import polynomial as poly
 from sklearn.base import BaseEstimator, TransformerMixin, OneToOneFeatureMixin
 from sklearn.utils.validation import check_is_fitted, validate_data
 
@@ -35,8 +36,14 @@ class IndexShift(TransformerMixin, OneToOneFeatureMixin, BaseEstimator):
         Transform the input data by shifting the spectrum.
     """
 
-    def __init__(self, shift: int = 0, random_state: Optional[int] = None):
+    def __init__(
+        self,
+        shift: int = 0,
+        fill_method: Literal["constant", "linear", "quadratic"] = "constant",
+        random_state: Optional[int] = None,
+    ):
         self.shift = shift
+        self.fill_method = fill_method
         self.random_state = random_state
 
     def fit(self, X: np.ndarray, y=None) -> "IndexShift":
@@ -111,10 +118,82 @@ class IndexShift(TransformerMixin, OneToOneFeatureMixin, BaseEstimator):
 
         # Calculate the standard normal variate
         for i, x in enumerate(X_):
-            X_[i] = self._shift_spectrum(x)
+            X_[i] = self._shift_vector(x)
 
         return X_.reshape(-1, 1) if X_.ndim == 1 else X_
 
     def _shift_spectrum(self, x) -> np.ndarray:
         shift_amount = self._rng.integers(-self.shift, self.shift, endpoint=True)
         return np.roll(x, shift_amount)
+
+    def _shift_vector(
+        self,
+        x: np.ndarray,
+    ) -> np.ndarray:
+        """
+        Shift vector with option to fill missing values.
+
+        Args:
+            arr: Input numpy array
+            shift: Number of positions to shift
+            fill_method: Method to fill missing values
+                'constant': fill with first/last value
+                'linear': fill using linear regression
+                'quadratic': fill using quadratic regression
+
+        Returns:
+            Shifted numpy array
+        """
+        shift = self._rng.integers(-self.shift, self.shift, endpoint=True)
+
+        result = np.roll(x, shift)
+
+        if self.fill_method == "constant":
+            if shift > 0:
+                result[:shift] = x[0]
+            elif shift < 0:
+                result[shift:] = x[-1]
+
+        elif self.fill_method == "linear":
+            if shift > 0:
+                x_ = np.arange(5)
+                coeffs = poly.polyfit(x_, x[:5], 1)
+
+                extrapolate_x = np.arange(-shift, 0)
+                extrapolated_values = poly.polyval(extrapolate_x, coeffs)
+
+                result[:shift] = extrapolated_values
+
+            elif shift < 0:
+                x_ = np.arange(5)
+                coeffs = poly.polyfit(x_, x[-5:], 1)
+
+                extrapolate_x = np.arange(len(x_), len(x_) - shift)
+                extrapolated_values = poly.polyval(extrapolate_x, coeffs)
+
+                result[shift:] = extrapolated_values
+
+        elif self.fill_method == "quadratic":
+            if shift > 0:
+                # Use first 3 values for quadratic regression
+                x_ = np.arange(5)
+                coeffs = poly.polyfit(x_, x[:5], 2)
+
+                # Extrapolate to fill shifted region
+                extrapolate_x = np.arange(-shift, 0)
+                extrapolated_values = poly.polyval(extrapolate_x, coeffs)
+
+                result[:shift] = extrapolated_values
+
+            elif shift < 0:
+                # Use last 3 values for quadratic regression
+                x_ = np.arange(5)
+                coeffs = poly.polyfit(x_, x[-5:], 2)
+
+                # Extrapolate to fill shifted region
+                extrapolate_x = np.arange(len(x_), len(x_) - shift)
+                extrapolated_values = poly.polyval(extrapolate_x, coeffs)
+
+                result[shift:] = extrapolated_values
+
+        return result

{chemotools-0.1.6 → chemotools-0.1.7}/chemotools/scatter/_extended_multiplicative_scatter_correction.py

@@ -1,4 +1,4 @@
-from typing import Optional
+from typing import Literal, Optional
 
 import numpy as np
 from sklearn.base import BaseEstimator, TransformerMixin, OneToOneFeatureMixin
@@ -46,18 +46,20 @@ class ExtendedMultiplicativeScatterCorrection(
     model-based pre-processing, doi:10.1016/j.chemolab.2021.104350
     """
 
+    ALLOWED_METHODS = ["mean", "median"]
+
+    # TODO: Check method is valid in instantiation. Right now it is check on fit because it breaks the scikitlearn check_estimator()
+
     def __init__(
         self,
-        reference: Optional[np.ndarray] = None,
-        use_mean: bool = True,
-        use_median: bool = False,
+        method: Literal["mean", "median"] = "mean",
         order: int = 2,
+        reference: Optional[np.ndarray] = None,
         weights: Optional[np.ndarray] = None,
     ):
-        self.reference = reference
-        self.use_mean = use_mean
-        self.use_median = use_median
+        self.method = method
         self.order = order
+        self.reference = reference
         self.weights = weights
 
     def fit(self, X: np.ndarray, y=None) -> "ExtendedMultiplicativeScatterCorrection":
@@ -104,21 +106,24 @@ class ExtendedMultiplicativeScatterCorrection(
             self.weights_ = np.array(self.weights)
             return self
 
-        if self.use_median:
-            self.reference_ = np.median(X, axis=0)
+        if self.method == "mean":
+            self.reference_ = X.mean(axis=0)
             self.indices_ = self._calculate_indices(X[0])
             self.A_ = self._calculate_A(self.indices_, self.reference_)
             self.weights_ = np.array(self.weights)
             return self
 
-        if self.use_mean:
-            self.reference_ = X.mean(axis=0)
+        elif self.method == "median":
+            self.reference_ = np.median(X, axis=0)
             self.indices_ = self._calculate_indices(X[0])
             self.A_ = self._calculate_A(self.indices_, self.reference_)
             self.weights_ = np.array(self.weights)
             return self
 
-        raise ValueError("No reference was provided")
+        else:
+            raise ValueError(
+                f"Invalid method: {self.method}. Must be one of {self.ALLOWED_METHODS}"
+            )
 
     def transform(self, X: np.ndarray, y=None) -> np.ndarray:
         """

{chemotools-0.1.6 → chemotools-0.1.7}/chemotools/scatter/_multiplicative_scatter_correction.py

@@ -1,4 +1,4 @@
-from typing import Optional
+from typing import Literal, Optional
 
 import numpy as np
 from sklearn.base import BaseEstimator, TransformerMixin, OneToOneFeatureMixin
@@ -37,16 +37,18 @@ class MultiplicativeScatterCorrection(
 
     """
 
+    ALLOWED_METHODS = ["mean", "median"]
+
+    # TODO: Check method is valid in instantiation. Right now it is check on fit because it breaks the scikitlearn check_estimator()
+
     def __init__(
         self,
+        method: Literal["mean", "median"] = "mean",
         reference: Optional[np.ndarray] = None,
-        use_mean: bool = True,
-        use_median: bool = False,
         weights: Optional[np.ndarray] = None,
     ):
+        self.method = method
         self.reference = reference
-        self.use_mean = use_mean
-        self.use_median = use_median
         self.weights = weights
 
     def fit(self, X: np.ndarray, y=None) -> "MultiplicativeScatterCorrection":
@@ -91,18 +93,23 @@ class MultiplicativeScatterCorrection(
             self.weights_ = np.array(self.weights)
             return self
 
-        if self.use_median:
-            self.reference_ = np.median(X, axis=0)
+        if self.method == "mean":
+            self.reference_ = X.mean(axis=0)
             self.A_ = self._calculate_A(self.reference_)
             self.weights_ = np.array(self.weights)
             return self
 
-        if self.use_mean:
-            self.reference_ = X.mean(axis=0)
+        elif self.method == "median":
+            self.reference_ = np.median(X, axis=0)
             self.A_ = self._calculate_A(self.reference_)
             self.weights_ = np.array(self.weights)
             return self
 
+        else:
+            raise ValueError(
+                f"Invalid method: {self.method}. Must be one of {self.ALLOWED_METHODS}"
+            )
+
         raise ValueError("No reference was provided")
 
     def transform(self, X: np.ndarray, y=None) -> np.ndarray:

{chemotools-0.1.6 → chemotools-0.1.7}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "chemotools"
-version = "0.1.6"
+version = "0.1.7"
 description = "chemotools: A Python Package that Integrates Chemometrics and scikit-learn"
 authors = ["Pau Cabaneros"]
 license = "MIT License"