feat(difficulty): implement Monte Carlo simulation for accurate difficulty calculation
This commit is contained in:
75
scripts/simulate_difficulty.js
Normal file
75
scripts/simulate_difficulty.js
Normal file
@@ -0,0 +1,75 @@
|
||||
|
||||
import fs from 'fs';
|
||||
import path from 'path';
|
||||
import { generateRandomGrid, calculateHints } from '../src/utils/puzzleUtils.js';
|
||||
import { solvePuzzle } from '../src/utils/solver.js';
|
||||
|
||||
const OUTPUT_FILE = 'difficulty_simulation_results.json';
|
||||
const CSV_FILE = 'difficulty_simulation_results.csv';
|
||||
|
||||
// Configuration
|
||||
const SIZES = [5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 60, 70, 80]; // Steps of 5 up to 50, then 10
|
||||
const DENSITIES = [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9];
|
||||
const SAMPLES_PER_POINT = 20; // Adjust based on time/accuracy needs
|
||||
|
||||
console.log('Starting Monte Carlo Simulation for Nonogram Difficulty...');
|
||||
console.log(`Config: Sizes=${SIZES.length}, Densities=${DENSITIES.length}, Samples=${SAMPLES_PER_POINT}`);
|
||||
|
||||
const results = [];
|
||||
const csvRows = ['size,density,avg_solved_percent,min_solved_percent,max_solved_percent,avg_time_ms'];
|
||||
|
||||
const startTime = Date.now();
|
||||
|
||||
for (const size of SIZES) {
|
||||
for (const density of DENSITIES) {
|
||||
let totalSolved = 0;
|
||||
let minSolved = 100;
|
||||
let maxSolved = 0;
|
||||
let totalTime = 0;
|
||||
|
||||
process.stdout.write(`Simulating Size: ${size}x${size}, Density: ${density} ... `);
|
||||
|
||||
for (let i = 0; i < SAMPLES_PER_POINT; i++) {
|
||||
const t0 = performance.now();
|
||||
|
||||
// 1. Generate
|
||||
const grid = generateRandomGrid(size, density);
|
||||
const { rowHints, colHints } = calculateHints(grid);
|
||||
|
||||
// 2. Solve
|
||||
const { percentSolved } = solvePuzzle(rowHints, colHints);
|
||||
|
||||
const t1 = performance.now();
|
||||
|
||||
totalSolved += percentSolved;
|
||||
minSolved = Math.min(minSolved, percentSolved);
|
||||
maxSolved = Math.max(maxSolved, percentSolved);
|
||||
totalTime += (t1 - t0);
|
||||
}
|
||||
|
||||
const avgSolved = totalSolved / SAMPLES_PER_POINT;
|
||||
const avgTime = totalTime / SAMPLES_PER_POINT;
|
||||
|
||||
results.push({
|
||||
size,
|
||||
density,
|
||||
avgSolved,
|
||||
minSolved,
|
||||
maxSolved,
|
||||
avgTime
|
||||
});
|
||||
|
||||
csvRows.push(`${size},${density},${avgSolved.toFixed(2)},${minSolved.toFixed(2)},${maxSolved.toFixed(2)},${avgTime.toFixed(2)}`);
|
||||
|
||||
console.log(`Avg Solved: ${avgSolved.toFixed(1)}%`);
|
||||
}
|
||||
}
|
||||
|
||||
const totalDuration = (Date.now() - startTime) / 1000;
|
||||
console.log(`Simulation complete in ${totalDuration.toFixed(1)}s`);
|
||||
|
||||
// Save results
|
||||
fs.writeFileSync(OUTPUT_FILE, JSON.stringify(results, null, 2));
|
||||
fs.writeFileSync(CSV_FILE, csvRows.join('\n'));
|
||||
|
||||
console.log(`Results saved to ${OUTPUT_FILE} and ${CSV_FILE}`);
|
||||
Reference in New Issue
Block a user