update to version 0.0.8 info in changelog

.gitignore

@@ -7,6 +7,7 @@ __pycache__/
 *.so
 
 # Distribution / packaging
+test.py
 .Python
 build/
 develop-eggs/

CHANGELOG.md

@@ -0,0 +1,16 @@
+## [0.0.8] - 2023-09-26
+### Added
+- **Vector Mean Calculation**: Added a new method to calculate the mean (average) of the vector's values. You can now use the `mean` method to obtain the mean value of a Vector.
+
+- **Vector Variance Calculation**: Implemented a method to compute the variance of the vector's values. You can now use the `variance` method to calculate the variance of a Vector.
+
+- **Vector Standard Deviation Calculation**: Introduced a method for calculating the standard deviation of the vector's values. You can now use the `std_deviation` method to find the standard deviation of a Vector.
+
+- **Linear Regression**: Added support for linear regression with the `LinearRegression` class. You can use the `fit` method to train a linear regression model and the `predict` method to make predictions based on the trained model.
+
+- **Time Series Analysis**: Introduced a new class, `TimeSeriesAnalysis`, for analyzing time series data.
+  - `moving_average(self, window_size: int) -> Vector`: Added a method for calculating the moving average of a time series with a specified window size.
+  - `exponential_smoothing(self, alpha: float) -> Vector`: Added a method for applying exponential smoothing to a time series with a given smoothing parameter.
+
+These new features provide additional statistical analysis capabilities for Vector objects, as well as the ability to perform linear regression on your data using the `LinearRegression` class.
+These new features enable users to perform time series analysis, including moving average and exponential smoothing, using the `TimeSeriesAnalysis` class. This enhances the capabilities of the library for working with time series data.

UNIMATHLIB/MathFunc/matrix.py

@@ -87,7 +87,8 @@ class Matrix():
         QR_eigen(max_iterations: int = 100):
             Compute eigenvalues and eigenvectors using QR algorithm
    
-    
+    Properties:
+        - shape: Return the shape (dimension) of the Matrix.
     Attributes:
         rows (int):
             The number of rows in the Matrix.

UNIMATHLIB/MathFunc/vector.py

@@ -35,7 +35,10 @@ class Vector:
         - cross(self, y): Calculate the cross product of two 3D Vectors.
         - size(self): Return the size (dimension) of the Vector.
         - _AllNum(vector): Check if all elements in the Vector are of type int or float.
-
+        - mean(self): Calculate the mean (average) of the vector's values.
+        - variance(self): Calculate the variance of the vector's values.
+        - std_deviation(self): Calculate the standard deviation of the vector's values.
+    
     Properties:
         - size: Return the size (dimension) of the Vector.
 
@@ -408,7 +411,7 @@ def clear(self):
 
 
 
-    def cross(self,y):
+    def cross(self,other):
         """Calculate the cross product of two 3-dimensional vectors.
 
         Args:
@@ -423,12 +426,12 @@ def cross(self,y):
         if isinstance(other, (list, tuple, set, dict, str)):
             other = Vector(other)
         if isinstance(other, Vector):
-            if len(self) != 3 or len(y) != 3:
+            if len(self) != 3 or len(other) != 3:
                 raise ValueError("Vector must be 3 dimensional")
             result = Vector([
-            self[1] * y[2] - self[2] * y[1],
-            self[2] * y[0] - self[0] * y[2],
-            self[0] * y[1] - self[1] * y[0]
+            self[1] * other[2] - self[2] * other[1],
+            self[2] * other[0] - self[0] * other[2],
+            self[0] * other[1] - self[1] * other[0]
             ])
             return result
 
@@ -459,6 +462,49 @@ def dist(self,y,p = 1):
         else:
             raise ValueError(f"Error in vector format, please check {(self,y)}")
 
+    def mean(self):
+        """
+        Calculate the mean (average) of the vector's values.
+
+        Raises:
+            ValueError: If the vector is empty.
+
+        Returns:
+            float: The mean value of the vector's elements.
+        """
+        if len(self) == 0:
+            raise ValueError("Empty Vector")
+
+        total = sum(self.vector)
+        return total / len(self.vector)
+
+    def variance(self):
+        """
+        Calculate the variance of the vector's values.
+
+        Raises:
+            ValueError: If there are fewer than 2 values in the vector.
+
+        Returns:
+            float: The variance of the vector's elements.
+        """
+        if len(self) < 2:
+            raise ValueError("Need min 2 values for variance")
+
+        mean_value = self.mean()
+        squared_diff = [(x - mean_value) ** 2 for x in self.vector]
+        return sum(squared_diff) / (len(self.vector) - 1)
+
+    def std_deviation(self):
+        """
+        Calculate the standard deviation of the vector's values.
+
+        Returns:
+            float: The standard deviation of the vector's elements.
+        """
+        return (self.variance())**0.5
+
+
     def _euclidean(self, y):
         """Calculate the Euclidean norm of two vectors.
 
@@ -470,6 +516,7 @@ def _euclidean(self, y):
         """
         return sum(abs(x-y)**2 for x,y in zip(self.vector,y.vector))**(1/2)
 
+
     def _manhattan(self, y):
         """Calculate the Manhattan norm of two vectors.
 

UNIMATHLIB/__init__.py

@@ -1,2 +1,4 @@
 from .MathFunc import *
-
+from .Linear import *
+from .timeSeriesAnalys import *
+from .fin import *

setup.py

@@ -1,7 +1,7 @@
 from setuptools import setup, find_packages
 
 setup(name='UniMathLib',
-      version='0.0.7',
+      version='0.0.8',
       license = "GNU GPLv3",
       description="""UniMathLib is a Python library that provides essential tools for 
       working with mathematical vectors and matrices. It simplifies tasks like vector 
@@ -44,7 +44,7 @@
 
       Identity and Diagonal Matrices: Create identity matrices of any size and diagonal matrices with specified diagonal values.
 
-
+      and others.
       UniMathLib simplifies complex mathematical tasks, making it a valuable tool for professionals and 
       students alike. Whether you need to perform basic vector operations or solve intricate linear systems,
       UniMathLib provides an efficient and user-friendly solution. Start using UniMathLib today to enhance your mathematical 

unimathlib/Linear/__init__.py

@@ -0,0 +1 @@
+from .linreg import *

unimathlib/Linear/linreg.py

@@ -0,0 +1,95 @@
+from ..MathFunc import Matrix, Vector
+import sys
+import io
+class LinearRegression:
+    """
+    Linear Regression Model.
+
+    This class implements a simple linear regression model using the method of least squares.
+
+    Attributes:
+        coefficients (Vector): The coefficients of the linear regression model, including the intercept.
+
+    Methods:
+        - __init__(self): Initialize an instance of the LinearRegression class.
+        - fit(self, X: Matrix, y: Matrix): Fit the linear regression model to the training data.
+        - predict(self, X: Matrix): Predict target values for new data.
+
+    Example:
+        # Create a LinearRegression instance
+        model = LinearRegression()
+
+        # Fit the model to training data
+        X_train = Matrix([[1], [2], [3], [4]])
+        y_train = Matrix([2, 4, 6, 8])
+        model.fit(X_train, y_train)
+
+        # Make predictions
+        X_test = Matrix([[5], [6]])
+        predictions = model.predict(X_test)
+    """
+    def __init__(self):
+        """
+        Initialize an instance of the LinearRegression class.
+
+        The coefficients attribute is set to None initially.
+        """
+        self.coefficients = None
+
+    def fit(self, X:Matrix, y:Matrix):
+        """
+        Fit the linear regression model to the training data.
+
+        Args:
+            X (Matrix): The feature matrix. Each row represents an observation.
+            y (Matrix): The target values.
+
+        Raises:
+            ValueError: If the lengths of X and y do not match.
+
+        Returns:
+            None
+        """
+        if not isinstance(X, Matrix) or not isinstance(y, Matrix): 
+            X = Matrix(X)
+            y = Matrix(y)
+        original_stdout = sys.stdout
+        sys.stdout = io.StringIO()
+        if len(X) != len(y):
+            raise ValueError("len(X) must be equal len(y)")
+
+        X_reg = Matrix(X.shape[0],X.shape[1]+1)
+        for i in range(len(X)):
+            n_r = Vector(X.shape[1]+1)
+            n_r[0], n_r[1:] = 1, X[i]
+            X_reg[i] = n_r
+        X_transpose = X_reg.transpose()
+        XTX = X_transpose*X_reg
+        XTY = X_transpose*y
+        self.coefficients = XTX.LU_solver(XTY)
+        sys.stdout = original_stdout
+
+    def predict(self, X):
+        """
+        Predict target values for new data.
+
+        Args:
+            X (Matrix): The feature matrix for prediction. Each row represents an observation.
+
+        Raises:
+            ValueError: If the model has not been trained (fit) yet.
+
+        Returns:
+            Vector: Predicted target values.
+        """
+        if self.coefficients is None:
+            raise ValueError("Use fit for model teach.")
+        if not isinstance(X, Matrix): 
+            X = Matrix(X)
+        X_reg = Matrix(X.shape[0],X.shape[1]+1)
+        for i in range(len(X)):
+            n_r = Vector(X.shape[1]+1)
+            n_r[0], n_r[1:] = 1, X[i]
+            X_reg[i] = n_r
+        return Vector([Vector(x) * Vector([el[0] for el in self.coefficients.matrix]) for x in X_reg])
+

unimathlib/fin/__init__.py

@@ -0,0 +1 @@
+from .finance import *

unimathlib/fin/finance.py

@@ -0,0 +1,83 @@
+from ..MathFunc import Vector
+
+class FinancialAnalysis:
+    """
+    Financial Analysis Class.
+
+    This class provides methods for financial analysis, including NPV and IRR calculations.
+
+    Methods:
+        - __init__(self): Initialize an instance of the FinancialAnalysis class.
+        - npv(self, cash_flows: Vector), discount_rate: float) -> float: Calculate the Net Present Value (NPV) of cash flows.
+        - irr(self, cash_flows: Vector) -> float: Calculate the Internal Rate of Return (IRR) of cash flows.
+
+    Example:
+        # Create a FinancialAnalysis instance
+        analyzer = FinancialAnalysis()
+
+        # Calculate NPV for cash flows with a discount rate of 0.1
+        cash_flows = [-100, 50, 60, 70]
+        discount_rate = 0.1
+        npv_result = analyzer.npv(cash_flows, discount_rate)
+
+        # Calculate IRR for cash flows
+        irr_result = analyzer.irr(cash_flows)
+    """
+
+    def __init__(self):
+        """
+        Initialize an instance of the FinancialAnalysis class.
+        """
+        pass
+
+    def npv(self, cash_flows, discount_rate: float) -> float:
+        """
+        Calculate the Net Present Value (NPV) of cash flows.
+
+        Args:
+            cash_flows (List[float]): List of cash flows over time.
+            discount_rate (float): The discount rate.
+
+        Returns:
+            float: The NPV value.
+        """
+        if not isinstance(cash_flows, Vector):
+            cash_flows = Vector(cash_flows)
+        npv_value = 0
+        for i, cash_flow in enumerate(cash_flows):
+            npv_value += cash_flow / (1 + discount_rate) ** i
+        return npv_value
+
+    def irr(self, cash_flows, max_iterations:int=1000, irr_guess:float = 0.1, tolerance:float=1e-6) -> float:
+        """
+        Calculate the Internal Rate of Return (IRR) of cash flows.
+
+        Args:
+            cash_flows (Vector): List of cash flows over time.
+            max_iterations (int, optional): Maximum number of iterations for IRR calculation. Defaults to 1000.
+            irr_guess (float, optional): Initial guess for IRR. Defaults to 0.1.
+            tolerance (float, optional): Tolerance level for convergence. Defaults to 1e-6.
+
+        Raises:
+            ValueError: Raised if the IRR calculation does not converge within the specified parameters.
+
+        Returns:
+            float: The calculated IRR value.
+        """
+        if not isinstance(cash_flows, Vector):
+            cash_flows = Vector(cash_flows)
+        for _ in range(max_iterations):
+            npv = self.npv(cash_flows, irr_guess)
+            npv_derivative = 0
+            for t, cash_flow in enumerate(cash_flows):
+                npv_derivative -= t * cash_flow / ((1 + irr_guess) ** (t + 1))
+
+            # Newton-Raphson step
+            irr_guess -= npv / npv_derivative
+
+            # Check for convergence
+            if abs(npv) < tolerance:
+                return irr_guess
+
+        # If no convergence is reached, return None or raise an exception
+        raise ValueError("IRR calculation did not converge")

unimathlib/timeSeriesAnalys/__init__.py

@@ -0,0 +1 @@
+from .tsy import *