PathFinder

PathFinder is a Python package that provides classic pathfinding algorithms such as:

• Breadth-First Search (BFS)

• Depth-First Search (DFS)

• Uniform Cost Search (UCS)

• A-Star Search (A*)

These algorithms work on 2D mazes represented as grids, and are designed to help with visualization, teaching, or solving pathfinding problems programmatically.

💾 Installation

pip install PathWise

📝 Features

• Supports 2D grid mazes with customizable start, goal, and wall positions

• Clear API to run any algorithm and get the path, cost, and visited nodes

• Easily extendable for diagonal movement or custom cost functions

• Suitable for AI projects, teaching, and maze-solving

⚖️ Algorithms Included

BFS

• Explores nodes level by level

• Guarantees shortest path if all moves have equal cost

DFS

• Explores deep into one branch before backtracking

• May not find the shortest path

UCS

• Uses a priority queue (cost-based)

• Always finds the lowest-cost path

A*

• Uses cost + heuristic (e.g. Manhattan distance)

• Highly efficient for large or complex mazes

💡 Usage

1. Representing the Maze

A maze is a 2D list of characters:

maze = [ ["S", " ", " ", "#", "G"], ["#", "#", " ", "#", " "], [" ", " ", " ", " ", " "], [" ", "#", "#", "#", " "], [" ", " ", " ", " ", " "] ]

S = Start

G = Goal

# = Wall (Represented by 1)

' ' = Open path (Represented by 0)

2. Running an Algorithm

import numpy as np
from PathWise import BFS, DFS, UCS, A_Star

Maze = [[...], [...], ...]
Maze = np.array(Maze).astype('str')

path, nodesExpanded = DFS(Maze, start, end)         #   or BFS
path, cost, nodesExpanded = UCS(Maze, start, end)   #   or A_Star

print("Path:", path)
print("Visited:", visited)
print("Cost: ", cost)                               #   if UCS or A_Star has been used

📁 Project Structure

./
├── PathWise/
│   ├── __init__.py
│   ├── A_Star.py
│   ├── BFS.py
│   ├── DFS.py
│   ├── Helper.py
│   └── UCS.py
├── .gitignore
├── LICENSE
├── README.md
└── setup.py

📃 License

This project is licensed under the MIT License.

Author

Developed by Mahdi Jaffery