Python/arithmetic_analysis/jacobi_iteration_method.py
Caeden 4d0c830d2c
Add flake8 pluin flake8 bugbear to pre-commit (#7132)
* ci(pre-commit): Add ``flake8-builtins`` additional dependency to ``pre-commit`` (#7104)

* refactor: Fix ``flake8-builtins`` (#7104)

* fix(lru_cache): Fix naming conventions in docstrings (#7104)

* ci(pre-commit): Order additional dependencies alphabetically (#7104)

* fix(lfu_cache): Correct function name in docstring (#7104)

* Update strings/snake_case_to_camel_pascal_case.py

Co-authored-by: Christian Clauss <cclauss@me.com>

* Update data_structures/stacks/next_greater_element.py

Co-authored-by: Christian Clauss <cclauss@me.com>

* Update digital_image_processing/index_calculation.py

Co-authored-by: Christian Clauss <cclauss@me.com>

* Update graphs/prim.py

Co-authored-by: Christian Clauss <cclauss@me.com>

* Update hashes/djb2.py

Co-authored-by: Christian Clauss <cclauss@me.com>

* refactor: Rename `_builtin` to `builtin_` ( #7104)

* fix: Rename all instances (#7104)

* refactor: Update variable names (#7104)

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* ci: Create ``tox.ini`` and ignore ``A003`` (#7123)

* revert: Remove function name changes (#7104)

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Rename tox.ini to .flake8

* Update data_structures/heap/heap.py

Co-authored-by: Dhruv Manilawala <dhruvmanila@gmail.com>

* refactor: Rename `next_` to `next_item` (#7104)

* ci(pre-commit): Add `flake8` plugin `flake8-bugbear` (#7127)

* refactor: Follow `flake8-bugbear` plugin (#7127)

* fix: Correct `knapsack` code (#7127)

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

Co-authored-by: Christian Clauss <cclauss@me.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Dhruv Manilawala <dhruvmanila@gmail.com>
2022-10-13 18:03:06 +02:00

168 lines
5.2 KiB
Python

"""
Jacobi Iteration Method - https://en.wikipedia.org/wiki/Jacobi_method
"""
from __future__ import annotations
import numpy as np
from numpy import float64
from numpy.typing import NDArray
# Method to find solution of system of linear equations
def jacobi_iteration_method(
coefficient_matrix: NDArray[float64],
constant_matrix: NDArray[float64],
init_val: list[int],
iterations: int,
) -> list[float]:
"""
Jacobi Iteration Method:
An iterative algorithm to determine the solutions of strictly diagonally dominant
system of linear equations
4x1 + x2 + x3 = 2
x1 + 5x2 + 2x3 = -6
x1 + 2x2 + 4x3 = -4
x_init = [0.5, -0.5 , -0.5]
Examples:
>>> coefficient = np.array([[4, 1, 1], [1, 5, 2], [1, 2, 4]])
>>> constant = np.array([[2], [-6], [-4]])
>>> init_val = [0.5, -0.5, -0.5]
>>> iterations = 3
>>> jacobi_iteration_method(coefficient, constant, init_val, iterations)
[0.909375, -1.14375, -0.7484375]
>>> coefficient = np.array([[4, 1, 1], [1, 5, 2]])
>>> constant = np.array([[2], [-6], [-4]])
>>> init_val = [0.5, -0.5, -0.5]
>>> iterations = 3
>>> jacobi_iteration_method(coefficient, constant, init_val, iterations)
Traceback (most recent call last):
...
ValueError: Coefficient matrix dimensions must be nxn but received 2x3
>>> coefficient = np.array([[4, 1, 1], [1, 5, 2], [1, 2, 4]])
>>> constant = np.array([[2], [-6]])
>>> init_val = [0.5, -0.5, -0.5]
>>> iterations = 3
>>> jacobi_iteration_method(coefficient, constant, init_val, iterations)
Traceback (most recent call last):
...
ValueError: Coefficient and constant matrices dimensions must be nxn and nx1 but
received 3x3 and 2x1
>>> coefficient = np.array([[4, 1, 1], [1, 5, 2], [1, 2, 4]])
>>> constant = np.array([[2], [-6], [-4]])
>>> init_val = [0.5, -0.5]
>>> iterations = 3
>>> jacobi_iteration_method(coefficient, constant, init_val, iterations)
Traceback (most recent call last):
...
ValueError: Number of initial values must be equal to number of rows in coefficient
matrix but received 2 and 3
>>> coefficient = np.array([[4, 1, 1], [1, 5, 2], [1, 2, 4]])
>>> constant = np.array([[2], [-6], [-4]])
>>> init_val = [0.5, -0.5, -0.5]
>>> iterations = 0
>>> jacobi_iteration_method(coefficient, constant, init_val, iterations)
Traceback (most recent call last):
...
ValueError: Iterations must be at least 1
"""
rows1, cols1 = coefficient_matrix.shape
rows2, cols2 = constant_matrix.shape
if rows1 != cols1:
raise ValueError(
f"Coefficient matrix dimensions must be nxn but received {rows1}x{cols1}"
)
if cols2 != 1:
raise ValueError(f"Constant matrix must be nx1 but received {rows2}x{cols2}")
if rows1 != rows2:
raise ValueError(
f"""Coefficient and constant matrices dimensions must be nxn and nx1 but
received {rows1}x{cols1} and {rows2}x{cols2}"""
)
if len(init_val) != rows1:
raise ValueError(
f"""Number of initial values must be equal to number of rows in coefficient
matrix but received {len(init_val)} and {rows1}"""
)
if iterations <= 0:
raise ValueError("Iterations must be at least 1")
table: NDArray[float64] = np.concatenate(
(coefficient_matrix, constant_matrix), axis=1
)
rows, cols = table.shape
strictly_diagonally_dominant(table)
# Iterates the whole matrix for given number of times
for _ in range(iterations):
new_val = []
for row in range(rows):
temp = 0
for col in range(cols):
if col == row:
denom = table[row][col]
elif col == cols - 1:
val = table[row][col]
else:
temp += (-1) * table[row][col] * init_val[col]
temp = (temp + val) / denom
new_val.append(temp)
init_val = new_val
return [float(i) for i in new_val]
# Checks if the given matrix is strictly diagonally dominant
def strictly_diagonally_dominant(table: NDArray[float64]) -> bool:
"""
>>> table = np.array([[4, 1, 1, 2], [1, 5, 2, -6], [1, 2, 4, -4]])
>>> strictly_diagonally_dominant(table)
True
>>> table = np.array([[4, 1, 1, 2], [1, 5, 2, -6], [1, 2, 3, -4]])
>>> strictly_diagonally_dominant(table)
Traceback (most recent call last):
...
ValueError: Coefficient matrix is not strictly diagonally dominant
"""
rows, cols = table.shape
is_diagonally_dominant = True
for i in range(0, rows):
total = 0
for j in range(0, cols - 1):
if i == j:
continue
else:
total += table[i][j]
if table[i][i] <= total:
raise ValueError("Coefficient matrix is not strictly diagonally dominant")
return is_diagonally_dominant
# Test Cases
if __name__ == "__main__":
import doctest
doctest.testmod()