n_queens_problem_go

N-Queens Problem (Go + Genetic Algorithm)

This project solves the N-Queens puzzle with a genetic algorithm.

• A chromosome is a permutation of row positions (one queen per column).
• Fitness is based on diagonal conflicts (lower is better, 0 is solved).
• The solver uses roulette selection, PMX crossover, optional mutation, and bounded population trimming.

Requirements

• Go 1.26+

Quick Start

go run .

Run with custom parameters:

go run . -size 8 -population 1000 -max-epochs 1000 -min-to-mate 5 -max-to-mate 20 -mutation-rate 0.05 -seed 1234

CLI Flags

Parameter Tuning Guide

Start with these presets, then adjust from there:

• Small boards (N=8 to N=16)
  • -population 500 to 2000
  • -max-epochs 500 to 2000
  • -min-to-mate 4 to 8
  • -max-to-mate 10 to 30
  • -mutation-rate 0.02 to 0.08
• Larger boards (N>=32)
  • -population 10000 to 50000
  • -max-epochs 3000 to 10000
  • -min-to-mate 10 to 30
  • -max-to-mate 30 to 120
  • -mutation-rate 0.03 to 0.12

Rules of thumb:

• If the best conflict score plateaus early, increase -mutation-rate slightly.
• If progress is noisy, reduce -mutation-rate slightly and increase -population.
• If runs are too slow, reduce -population first, then -max-epochs.
• Keep -max-to-mate noticeably higher than -min-to-mate for diversity.
• For difficult sizes, run multiple seeds and keep the best result.

Example commands:

# Fast local run for 8x8
go run . -size 8 -population 1000 -max-epochs 1200 -min-to-mate 5 -max-to-mate 20 -mutation-rate 0.05 -seed 42

# Stronger run for 32x32
go run . -size 32 -population 20000 -max-epochs 6000 -min-to-mate 15 -max-to-mate 80 -mutation-rate 0.06 -seed 42

Reproducible Runs

Use a fixed -seed to make runs deterministic.

go run . -size 14 -seed 2026

The configured seed and mutation rate are written to logs at startup.

Output

• The terminal prints the final board.
• Each queen cell shows its conflict count (00 means no conflicts).
• Structured logs are written to app.log.

Example solved 8x8 board:

╔════╤════╤════╤════╤════╤════╤════╤════╗
║    │    │    │    │    │ 00 │    │    ║
╟────┼────┼────┼────┼────┼────┼────┼────╢
║ 00 │    │    │    │    │    │    │    ║
╟────┼────┼────┼────┼────┼────┼────┼────╢
║    │    │    │    │ 00 │    │    │    ║
╟────┼────┼────┼────┼────┼────┼────┼────╢
║    │ 00 │    │    │    │    │    │    ║
╟────┼────┼────┼────┼────┼────┼────┼────╢
║    │    │    │    │    │    │    │ 00 ║
╟────┼────┼────┼────┼────┼────┼────┼────╢
║    │    │ 00 │    │    │    │    │    ║
╟────┼────┼────┼────┼────┼────┼────┼────╢
║    │    │    │    │    │    │ 00 │    ║
╟────┼────┼────┼────┼────┼────┼────┼────╢
║    │    │    │ 00 │    │    │    │    ║
╚════╧════╧════╧════╧════╧════╧════╧════╝

Development

Run tests:

go test ./...

Run benchmarks:

go test -run '^$' -bench PMX -benchmem ./...
go test -run '^$' -bench CountConflicts -benchmem ./...