feat: Add general interface StatisticalProcedure

* added base StatisticalProcedure protocol
* replaced NDArray[np.float64] with np.ndarray

Author: Ангелина

src/procedures/semiparametric/nm_semi_param_algorithms/g_estimation_convolution.py

@@ -11,7 +11,6 @@
 BOHMAN_DELTA_DEFAULT_VALUE: float = 0.0001
 X_DATA_DEFAULT_VALUE: List[float] = [1.0]
 
-
 class NMEstimationDensityInvMT:
     """Estimation of mixing density function g (xi density function) of NM mixture
     represented in canonical form Y = xi + sigma*N.
@@ -34,8 +33,8 @@ class ParamsAnnotation(TypedDict, total=False):
         bohman_n: int
         bohman_delta: float
 
-    def __init__(self, sample: Optional[_typing.NDArray[np.float64]] = None, **kwargs: Unpack[ParamsAnnotation]):
-        self.sample: _typing.NDArray[np.float64] = np.array([]) if sample is None else sample
+    def __init__(self, sample: Optional[np.ndarray] = None, **kwargs: Unpack[ParamsAnnotation]):
+        self.sample: np.ndarray = np.array([]) if sample is None else sample
         self.n: int = len(self.sample)
         (
             self.x_data,
@@ -93,7 +92,7 @@ def cdf(self, X: np.ndarray) -> np.ndarray:
 
             return F_x.real
 
-    def compute(self, sample: _typing.NDArray[np.float64]) -> EstimateResult:
+    def compute(self, sample: np.ndarray) -> EstimateResult:
         inv = self.BohmanA(N=self.bohman_n, delta=self.bohman_delta)
         inv.fit(self.characteristic_function_xi)
         x_data_array = np.array(self.x_data, dtype=np.float64)

src/procedures/semiparametric/nv_semi_param_algorithms/g_estimation_given_mu.py

@@ -49,13 +49,13 @@ class ParamsAnnotation(TypedDict, total=False):
         integration_tolerance: float
         integration_limit: int
 
-    def __init__(self, sample: Optional[_typing.NDArray[np.float64]] = None, **kwargs: Unpack[ParamsAnnotation]):
+    def __init__(self, sample: Optional[np.ndarray] = None, **kwargs: Unpack[ParamsAnnotation]):
         self.x_powers: Dict[float, np.ndarray] = {}
         self.second_u_integrals: np.ndarray
         self.first_u_integrals: np.ndarray
         self.gamma_grid: np.ndarray
         self.v_grid: np.ndarray
-        self.sample: _typing.NDArray[np.float64] = np.array([]) if sample is None else sample
+        self.sample: np.ndarray = np.array([]) if sample is None else sample
         self.n: int = len(self.sample)
         (
             self.gmm,
@@ -173,6 +173,6 @@ def compute_integrals_for_x(self, x: float) -> float:
         total = np.sum(first_integral + second_integral) / self.denominator
         return max(0.0, total.real)
 
-    def compute(self, sample: _typing.NDArray[np.float64]) -> EstimateResult:
+    def compute(self, sample: np.ndarray) -> EstimateResult:
         y_data = [self.compute_integrals_for_x(x) for x in self.x_data]
         return EstimateResult(list_value=y_data, success=True)

src/procedures/semiparametric/nv_semi_param_algorithms/g_estimation_post_widder.py

@@ -77,7 +77,7 @@ def p_x_estimation(self, x: float) -> float:
         )
         return result.real
 
-    def compute(self, sample: np._typing.NDArray) -> EstimateResult:
+    def compute(self, sample: np.ndarray) -> EstimateResult:
         """Estimate g(x)
 
         Args:

src/procedures/semiparametric/nvm_semi_param_algorithms/g_estimation_given_mu.py

@@ -51,13 +51,13 @@ class ParamsAnnotation(TypedDict, total=False):
         integration_tolerance: float
         integration_limit: int
 
-    def __init__(self, sample: Optional[_typing.NDArray[np.float64]] = None, **kwargs: Unpack[ParamsAnnotation]):
+    def __init__(self, sample: Optional[np.ndarray] = None, **kwargs: Unpack[ParamsAnnotation]):
         self.x_powers: Dict[float, np.ndarray] = {}
         self.second_u_integrals: np.ndarray
         self.first_u_integrals: np.ndarray
         self.gamma_grid: np.ndarray
         self.v_grid: np.ndarray
-        self.sample: _typing.NDArray[np.float64] = np.array([]) if sample is None else sample
+        self.sample: np.ndarray= np.array([]) if sample is None else sample
         self.n: int = len(self.sample)
         (
             self.mu,
@@ -179,6 +179,6 @@ def compute_integrals_for_x(self, x: float) -> float:
         total = np.sum(first_integral + second_integral) / self.denominator
         return max(0.0, total.real)
 
-    def compute(self, sample: _typing.NDArray[np.float64]) -> EstimateResult:
+    def compute(self, sample: np.ndarray) -> EstimateResult:
         y_data = [self.compute_integrals_for_x(x) for x in self.x_data]
         return EstimateResult(list_value=y_data, success=True)

src/procedures/semiparametric/nvm_semi_param_algorithms/g_estimation_given_mu_rqmc_based.py

@@ -52,13 +52,13 @@ class ParamsAnnotation(TypedDict, total=False):
         integration_tolerance: float
         integration_limit: int
 
-    def __init__(self, sample: Optional[_typing.NDArray[np.float64]] = None, **kwargs: Unpack[ParamsAnnotation]):
+    def __init__(self, sample: Optional[np.ndarray] = None, **kwargs: Unpack[ParamsAnnotation]):
         self.x_powers: Dict[float, np.ndarray] = {}
         self.second_u_integrals: np.ndarray
         self.first_u_integrals: np.ndarray
         self.gamma_grid: np.ndarray
         self.v_grid: np.ndarray
-        self.sample: _typing.NDArray[np.float64] = np.array([]) if sample is None else sample
+        self.sample: np.ndarray = np.array([]) if sample is None else sample
         self.n: int = len(self.sample)
         (
             self.mu,
@@ -171,6 +171,6 @@ def compute_integrals_for_x(self, x: float) -> float:
         total = (first_integral + second_integral) / self.denominator
         return max(0.0, total.real)
 
-    def compute(self, sample: np._typing.NDArray) -> EstimateResult:
+    def compute(self, sample: np.ndarray) -> EstimateResult:
         y_data = [self.compute_integrals_for_x(x) for x in self.x_data]
         return EstimateResult(list_value=y_data, success=True)

src/procedures/semiparametric/nvm_semi_param_algorithms/mu_estimation.py

@@ -84,7 +84,7 @@ def __w(self, p: float, sample: np._typing.NDArray) -> float:
             y += e * self.omega(x)
         return y
 
-    def compute(self, sample: np._typing.NDArray) -> EstimateResult:
+    def compute(self, sample: np.ndarray) -> EstimateResult:
         """Root of this function is an estimation of mu
 
         Args:

src/procedures/semiparametric/statistical_procedure.py

@@ -0,0 +1,12 @@
+from typing import Protocol
+import numpy as np
+
+from src.estimators.estimate_result import EstimateResult
+
+class StatisticalProcedure(Protocol):
+    """Protocol for statistical algorithms that compute estimates from data.
+
+    """
+    def compute(self, sample: np.ndarray) -> EstimateResult:
+        pass
+