Skip to content
/ aif Public

nextdataAI: a Python library for solving mazes using classical algorithms (A*, BFS, DFS, Dijkstra, Greedy), reinforcement learning (QLearning, QTable, QNN, QLSTM), and novel pseudo-heuristics. Features include a neural network-based heuristic approximating Manhattan distance (NNHeuristic) and a custom SManhattan heuristic, a dataset gener

Notifications You must be signed in to change notification settings

nextdataAI/aif

Repository files navigation

Artificial Intelligence Foundamentals

pip install nextdataAI 

wakatime

  • Generate some Mazes
  • Implement course solutions
    • BFS
    • DFS
    • Greedy
    • A*
      • Manhattan Heuristic
      • Euclidean Heuristic
      • Overkill Heuristic that destroys both Manhattan and Euclidean
    • Dijkstra
    • Genetic Algorithm
  • DL/RL approaches
    • Q-Learning
      • Q-Learning with NN and Experience Replay
      • Q-Learning with LSTM and Experience Replay
    • A* with NN heuristic
    • A* with LSTM heuristic
    • A* with CNN heuristic
  • Evaluation Metrics
    • Time
    • Number of Nodes Expanded
    • Number of Nodes in the Path
    • Number of Nodes in the Solution
    • Number of Nodes in the Maze
Heuristics Manhattan Euclidean NNHeuristic
Implemented

About

nextdataAI: a Python library for solving mazes using classical algorithms (A*, BFS, DFS, Dijkstra, Greedy), reinforcement learning (QLearning, QTable, QNN, QLSTM), and novel pseudo-heuristics. Features include a neural network-based heuristic approximating Manhattan distance (NNHeuristic) and a custom SManhattan heuristic, a dataset gener

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •