File size: 23,815 Bytes
dd757c3 a62ac7b eb24714 a62ac7b dd757c3 a62ac7b ef73f57 eb24714 a62ac7b eb24714 a62ac7b eb24714 dd757c3 a62ac7b ef73f57 a62ac7b eb24714 a62ac7b ef73f57 eb24714 ef73f57 eb24714 ef73f57 eb24714 ef73f57 eb24714 ef73f57 d79ded4 eb24714 d79ded4 eb24714 d79ded4 dd757c3 eb24714 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 |
{% extends "layout.html" %}
{% block content %}<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<title>🔬 Visual Random Forest Classifier (2D)</title>
<script src="https://cdn.plot.ly/plotly-2.32.0.min.js"></script>
<script src="https://cdn.tailwindcss.com"></script>
<script src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-mml-chtml.js"></script>
<style>
/* Basic styling for info icons and tooltips */
.info-icon {
cursor: help;
margin-left: 5px;
color: #6B7280; /* gray-500 */
position: relative; /* Needed for tooltip positioning */
display: inline-block;
}
.tooltip {
visibility: hidden;
width: 250px; /* Wider for more detail */
background-color: #333;
color: #fff;
text-align: center;
border-radius: 6px;
padding: 8px 10px; /* More padding */
position: absolute;
z-index: 10; /* Above other elements */
bottom: 125%; /* Position above the icon */
left: 50%;
margin-left: -125px; /* Center the tooltip */
opacity: 0;
transition: opacity 0.3s;
font-size: 0.85rem; /* Slightly smaller font for tooltips */
line-height: 1.4;
}
.info-icon:hover .tooltip {
visibility: visible;
opacity: 1;
}
/* Triangle for tooltip (optional) */
.tooltip::after {
content: "";
position: absolute;
top: 100%;
left: 50%;
margin-left: -5px;
border-width: 5px;
border-style: solid;
border-color: #333 transparent transparent transparent;
}
/* Styling for highlighting text in explanations */
.highlight-blue { color: #2563EB; font-weight: 600; } /* Tailwind blue-600 */
.highlight-red { color: #DC2626; font-weight: 600; } /* Tailwind red-600 */
.highlight-green { color: #16A34A; font-weight: 600; } /* Tailwind green-600 */
.highlight-bold { font-weight: 600; }
/* Styles for the flow chart boxes (reusing from KNN) */
.flow-box {
background-color: #F3F4F6; /* gray-100 */
border-radius: 0.5rem; /* rounded-lg */
padding: 1.5rem; /* p-6 */
text-align: center;
box-shadow: 0 4px 6px -1px rgba(0, 0, 0, 0.1), 0 2px 4px -1px rgba(0, 0, 0, 0.06); /* shadow-md */
min-height: 120px; /* Ensure boxes have some height */
display: flex;
flex-direction: column;
justify-content: center;
align-items: center;
}
.flow-arrow {
font-size: 2.5rem;
color: #9CA3AF; /* gray-400 */
margin: 0 1rem;
display: flex;
align-items: center;
justify-content: center;
}
</style>
</head>
<body class="bg-gray-100 text-gray-900">
<div class="max-w-5xl mx-auto mt-10 bg-white p-8 rounded-xl shadow-lg">
<h1 class="text-3xl font-bold mb-4 text-center">🔬 Visual Random Forest Classifier (2D)</h1>
<p class="mb-6 text-center text-gray-600">
An ensemble learning method for classification that operates by constructing a multitude of decision trees.
</p>
<div class="grid grid-cols-1 md:grid-cols-2 gap-6 mb-6">
<div>
<label for="testX" class="block font-medium mb-1 flex items-center">
Test Point X1:
<span class="info-icon">
ⓘ
<span class="tooltip">
The <span class="highlight-bold">X1-coordinate</span> of the new data point (<span class="highlight-green">green 'x'</span>) that you want the Random Forest to classify.
</span>
</span>
</label>
<input type="number" id="testX" value="4" class="w-24 px-2 py-1 border rounded" onchange="predict()">
</div>
<div>
<label for="testY" class="block font-medium mb-1 flex items-center">
Test Point X2:
<span class="info-icon">
ⓘ
<span class="tooltip">
The <span class="highlight-bold">X2-coordinate</span> of the new data point (<span class="highlight-green">green 'x'</span>) that you want the Random Forest to classify.
</span>
</span>
</label>
<input type="number" id="testY" value="2" class="w-24 px-2 py-1 border rounded" onchange="predict()">
</div>
</div>
<div class="mb-4 text-center">
<button onclick="predict()" class="bg-blue-500 hover:bg-blue-600 text-white px-6 py-2 rounded-lg text-lg transition duration-200">
Run Random Forest Prediction
</button>
</div>
<div id="plot" class="border rounded-lg shadow-inner h-[400px] md:h-[500px] w-full"></div>
<p id="result" class="mt-4 font-bold text-lg text-center text-gray-800"></p>
<div class="mt-10 p-6 bg-purple-50 rounded-xl border border-purple-200">
<h2 class="text-2xl font-bold mb-6 text-center text-purple-700">How Random Forest Classifies Your Data</h2>
<div class="flex flex-wrap justify-center items-center gap-4">
<div class="flow-box bg-purple-100">
<span class="text-5xl mb-2">🌳</span>
<p class="text-lg font-semibold text-purple-800">Many Decision Trees</p>
<p class="text-sm text-purple-600">Each trained on a random subset</p>
</div>
<div class="flow-arrow">→</div>
<div class="flow-box bg-purple-100">
<span class="text-5xl mb-2">📍</span>
<p class="text-lg font-semibold text-purple-800">New Data Point</p>
<p class="text-sm text-purple-600">Sent to ALL trees</p>
</div>
<div class="flow-arrow">→</div>
<div class="flow-box bg-purple-100">
<span class="text-5xl mb-2">📊</span>
<p class="text-lg font-semibold text-purple-800">Individual Predictions</p>
<p class="text-sm text-purple-600">Each tree "votes" on the class</p>
</div>
<div class="flow-arrow block md:hidden">↓</div>
<div class="flow-arrow hidden md:block">→</div>
<div class="flow-box bg-purple-100">
<span class="text-5xl mb-2">🗳️</span>
<p class="text-lg font-semibold text-purple-800">Majority Vote</p>
<p class="text-sm text-purple-600">Most common prediction wins</p>
</div>
<div class="flow-arrow">→</div>
<div class="flow-box bg-purple-100">
<span class="text-5xl mb-2">✅</span>
<p class="text-lg font-semibold text-purple-800">Final Classification</p>
<p class="text-sm text-purple-600">Robust and accurate</p>
</div>
</div>
<p class="mt-6 text-center text-gray-600 text-sm">
Random Forest combines the power of many individual decision trees to make a more robust and accurate classification, leveraging collective intelligence.
</p>
</div>
<div class="mt-8 text-center">
<a href="/liar" class="bg-purple-600 text-white px-4 py-2 rounded hover:bg-purple-700">
🔍 Go to Liar Predictor
</a>
<div class="mt-8 p-6 bg-gray-50 rounded-lg border border-gray-200">
<h2 class="text-2xl font-bold mb-4 text-center text-blue-700">Understanding Random Forest</h2>
<p class="mb-4 text-gray-700">
Random Forest is an <span class="highlight-bold">ensemble learning method</span> that builds a "forest" of decision trees. For classification tasks, it outputs the class that is the mode of the classes (majority vote) of the individual trees. It's known for its high accuracy and ability to handle complex datasets.
</p>
<h3 class="text-xl font-semibold mb-2">Key Concepts:</h3>
<ul class="list-disc list-inside text-gray-700 mb-4">
<li class="mb-2">
<span class="highlight-bold">Ensemble Learning:</span> Instead of relying on a single model, Random Forest combines predictions from multiple models (decision trees) to improve overall accuracy and robustness.
</li>
<li class="mb-2">
<span class="highlight-bold">Decision Trees:</span> Each tree in the forest makes a prediction independently. A single decision tree creates axis-parallel splits, leading to rectangular decision regions.
</li>
<li class="mb-2">
<span class="highlight-bold">Randomness:</span> Random Forest introduces randomness in two ways:
<ol class="list-decimal list-inside ml-4">
<li><span class="highlight-bold">Bagging (Bootstrap Aggregating):</span> Each tree is trained on a random subset of the training data (with replacement).</li>
<li><span class="highlight-bold">Feature Randomness:</span> When splitting a node, each tree considers only a random subset of the available features. This decorrelates the trees.</li>
</ol>
</li>
<li class="mb-2">
<span class="highlight-bold">Decision Boundary:</span> Unlike a single decision tree's sharp, rectangular boundaries, the Random Forest's decision boundary is the aggregated result of many trees. This often results in a smoother, more complex, and often non-linear boundary, as seen in the plot.
</li>
</ul>
<h3 class="text-xl font-semibold mb-2">How this Visualization Works:</h3>
<ul class="list-disc list-inside text-gray-700 mb-4">
<li class="mb-2">
<span class="highlight-red">Class 1 (Red Circles):</span> These are your labeled data points belonging to Class 1.
</li>
<li class="mb-2">
<span class="highlight-blue">Class 0 (Blue Circles):</span> These are your labeled data points belonging to Class 0.
</li>
<li class="mb-2">
<span class="highlight-green">Test Point (Green 'x'):</span> This is the new, unlabeled data point you want to classify. You can adjust its X1 and X2 coordinates.
</li>
<li class="mb-2">
<span class="highlight-bold">Colored Background:</span> This represents the <span class="highlight-bold">decision boundary</span> of the trained Random Forest model.
<ul>
<li><span class="highlight-red">Red regions</span> indicate areas where the Random Forest predicts Class 1.</li>
<li><span class="highlight-blue">Blue regions</span> indicate areas where the Random Forest predicts Class 0.</li>
</ul>
The smoothness and complexity of this boundary are a result of the ensemble nature of Random Forest.
</li>
</ul>
<p class="mt-4 text-sm text-gray-600">
*The plot will show the decision boundary by predicting the class for a grid of points covering the entire plot area. The color of each grid point reflects the predicted class, creating the background regions.*
</p>
</div>
</div>
<script>
// Sample labeled points (training data) - you can adjust these
// Using distinct clusters for better visualization of decision boundary
let labeledPoints = [
// Class 0 (Blue)
[1.0, 1.0, 0], [1.5, 1.8, 0], [2.0, 1.2, 0], [1.2, 2.5, 0], [2.5, 2.0, 0],
[3.0, 1.0, 0], [3.5, 2.2, 0], [2.8, 0.5, 0], [1.8, 0.8, 0], [0.5, 2.0, 0],
[0.8, 3.0, 0], [2.2, 3.5, 0], [3.2, 3.0, 0], [4.0, 2.5, 0], [4.5, 1.8, 0],
// Class 1 (Red)
[5.0, 5.0, 1], [5.5, 4.2, 1], [6.0, 5.8, 1], [4.8, 6.0, 1], [6.2, 6.5, 1],
[7.0, 5.0, 1], [6.5, 4.0, 1], [5.8, 6.8, 1], [7.2, 6.0, 1], [7.8, 5.5, 1],
[8.0, 7.0, 1], [7.5, 7.5, 1], [6.8, 7.2, 1], [5.2, 7.8, 1], [4.0, 6.0, 1]
];
function getTestPoint() {
const testXInput = document.getElementById('testX');
const testYInput = document.getElementById('testY');
const testX = parseFloat(testXInput.value);
const testY = parseFloat(testYInput.value);
if (isNaN(testX) || isNaN(testY)) {
document.getElementById("result").innerText = "❌ Please enter valid numbers for Test Point X1 and X2.";
testXInput.value = 4; // Reset to default
testYInput.value = 2; // Reset to default
return null;
}
return [testX, testY];
}
async function predict() {
const testPoint = getTestPoint();
if (testPoint === null) {
return; // Stop if testPoint input is invalid
}
document.getElementById("result").innerText = "Calculating decision boundary and prediction...";
console.log("Predicting with test point =", testPoint);
// *** API Call to your Python Backend ***
const res = await fetch("/rf_visual_predict", {
method: "POST",
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({
points: labeledPoints,
test_point: testPoint
// You could add n_estimators, max_depth here if you want them interactive
})
});
if (!res.ok) {
document.getElementById("result").innerText = `❌ Error from backend: ${res.statusText}. Ensure your Flask backend is running.`;
console.error("Backend error:", res.statusText);
return;
}
const data = await res.json();
console.log("Response:", data);
document.getElementById("result").innerText = `✅ Predicted Class for (${testPoint[0]}, ${testPoint[1]}): ${data.prediction == 0 ? 'Class 0 (Blue)' : 'Class 1 (Red)'}`;
// Prepare traces for Plotly
const class0 = labeledPoints.filter(p => p[2] == 0);
const class1 = labeledPoints.filter(p => p[2] == 1);
const trace0 = {
x: class0.map(p => p[0]),
y: class0.map(p => p[1]),
mode: 'markers',
name: 'Class 0',
marker: { color: 'blue', size: 10, symbol: 'circle' }
};
const trace1 = {
x: class1.map(p => p[0]),
y: class1.map(p => p[1]),
mode: 'markers',
name: 'Class 1',
marker: { color: 'red', size: 10, symbol: 'circle' }
};
const testTrace = {
x: [testPoint[0]],
y: [testPoint[1]],
mode: 'markers',
name: 'Test Point',
marker: { color: 'green', size: 14, symbol: 'x' }
};
// Decision Boundary (Contour/Heatmap)
// The 'z' data comes from the backend, representing predictions on a grid
const boundaryTrace = {
z: data.decision_boundary_z,
x: data.decision_boundary_x_coords, // X coordinates of the grid
y: data.decision_boundary_y_coords, // Y coordinates of the grid
type: 'contour',
colorscale: [
['0.0', 'rgba(0, 0, 255, 0.2)'], // Light blue for Class 0
['1.0', 'rgba(255, 0, 0, 0.2)'] // Light red for Class 1
],
showscale: false, // No color bar needed
hoverinfo: 'skip', // Don't show hover details for the boundary
opacity: 0.8 // Make it slightly transparent
};
// Plotly layout and redraw
Plotly.newPlot('plot', [boundaryTrace, trace0, trace1, testTrace], {
title: `Random Forest Decision Boundary`,
xaxis: { title: 'X1', range: [0, 9] }, // Adjust range based on your data
yaxis: { title: 'X2', range: [0, 9] }, // Adjust range
autosize: true, // Auto resize
hovermode: 'closest',
margin: { t: 40, b: 40, l: 40, r: 10 } // Clean margins
}, { responsive: true });
}
window.onload = () => predict();
</script>
<div class="mt-12 bg-white rounded-lg shadow-lg p-6 border border-gray-300">
<h2 class="text-2xl font-bold text-center text-purple-700 mb-4">🌳 Single Tree vs Random Forest</h2>
<div id="treeComparisonPlot" class="w-full h-[800px] md:h-[500px]"></div>
</div>
<script>
// This function detects screen size and returns the appropriate layout
function getTreeLayout() {
const isMobile = window.innerWidth < 768; // Standard mobile breakpoint
const singleTreeTrace = {
type: "scatter",
mode: "markers+lines+text",
x: [2, 1, 3, 0.5, 1.5, 2.5, 3.5],
y: [3, 2, 2, 1, 1, 1, 1],
text: ["Root", "", "", "class 1", "class 1", "class 2", "class 2"],
textposition: "top center",
marker: { size: isMobile ? 20 : 30, color: "royalblue" }, // Smaller dots on mobile
line: { color: 'royalblue', width: 2 },
name: "Single Decision Tree",
showlegend: false,
xaxis: 'x1',
yaxis: 'y1'
};
const forestTraces = [
{
type: "scatter",
mode: "markers+lines+text",
x: [7, 6.5, 7.5],
y: [3, 2, 2],
text: ["Tree 1", "class 1", "class 1"],
textposition: "top center",
marker: { size: isMobile ? 18 : 28, color: "red" },
line: { color: 'red', width: 2 },
showlegend: false,
xaxis: 'x2',
yaxis: 'y2'
},
{
type: "scatter",
mode: "markers+lines+text",
x: [9, 8.5, 9.5],
y: [3, 2, 2],
text: ["Tree 2", "class 2", "class 2"],
textposition: "top center",
marker: { size: isMobile ? 18 : 28, color: "green" },
line: { color: 'green', width: 2 },
showlegend: false,
xaxis: 'x2',
yaxis: 'y2'
},
{
type: "scatter",
mode: "markers+lines+text",
x: [11, 10.5, 11.5],
y: [3, 2, 2],
text: ["Tree 3", "class 3", "class 3"],
textposition: "top center",
marker: { size: isMobile ? 18 : 28, color: "orange" },
line: { color: 'orange', width: 2 },
showlegend: false,
xaxis: 'x2',
yaxis: 'y2'
}
];
let layout;
if (isMobile) {
// *** MOBILE LAYOUT: Stacked Vertically ***
layout = {
grid: { rows: 2, columns: 1, pattern: 'independent' },
// Top Graph (Single Tree)
xaxis: {
title: { text: "Single Decision Tree", font: { size: 14 } },
showgrid: false, zeroline: false
},
yaxis: { domain: [0.55, 1], showgrid: false, zeroline: false },
// Bottom Graph (Random Forest)
xaxis2: {
title: { text: "Random Forest (Multiple Trees)", font: { size: 14 } },
showgrid: false, zeroline: false
},
yaxis2: { domain: [0, 0.45], showgrid: false, zeroline: false },
showlegend: false,
margin: { t: 40, b: 40, l: 20, r: 20 },
autosize: true,
title: { text: "Visualizing Decision Tree vs Random Forest", font: { size: 16 } }
};
} else {
// *** DESKTOP LAYOUT: Side-by-Side ***
layout = {
grid: { rows: 1, columns: 2, pattern: 'independent' },
// Left Graph
xaxis: {
domain: [0, 0.45],
title: { text: "Single Decision Tree", font: { size: 14 } },
showgrid: false, zeroline: false
},
yaxis: { domain: [0, 1], showgrid: false, zeroline: false },
// Right Graph
xaxis2: {
domain: [0.55, 1],
title: { text: "Random Forest (Multiple Trees)", font: { size: 14 } },
showgrid: false, zeroline: false
},
yaxis2: { domain: [0, 1], showgrid: false, zeroline: false },
showlegend: false,
margin: { t: 50, b: 40 },
autosize: true,
title: { text: "Visualizing Decision Tree vs Random Forest", font: { size: 18 } }
};
}
return { traces: [singleTreeTrace, ...forestTraces], layout: layout };
}
function drawTreePlot() {
const data = getTreeLayout();
Plotly.react('treeComparisonPlot', data.traces, data.layout, {responsive: true});
}
// Initial draw
drawTreePlot();
// Redraw on resize to switch layouts
window.addEventListener('resize', drawTreePlot);
</script>
<div class="mt-8 text-gray-800 space-y-4">
<h3 class="text-xl font-bold text-blue-700">🔍 Working of Random Forest Algorithm</h3>
<ul class="list-disc list-inside space-y-2">
<li><strong>Create Many Decision Trees:</strong> The algorithm makes many decision trees using different random parts of the data.</li>
<li><strong>Pick Random Features:</strong> Each tree picks a random subset of features to make splits. This keeps trees diverse.</li>
<li><strong>Each Tree Makes a Prediction:</strong> Every tree gives its own output.</li>
<li><strong>Combine the Predictions:</strong>
<ul class="ml-6 list-disc">
<li><em>Classification:</em> Uses majority voting across trees.</li>
<li><em>Regression:</em> Averages the outputs of all trees.</li>
</ul>
</li>
<li><strong>Why It Works:</strong> Randomness prevents overfitting and improves overall prediction accuracy.</li>
</ul>
<h3 class="text-xl font-bold text-blue-700 mt-6">🌟 Key Features of Random Forest</h3>
<ul class="list-disc list-inside space-y-2">
<li><strong>Handles Missing Data:</strong> Works even with some missing values.</li>
<li><strong>Shows Feature Importance:</strong> Identifies most important features for prediction.</li>
<li><strong>Handles Complex Data:</strong> Efficient with large datasets and many features.</li>
<li><strong>Versatile:</strong> Works for both classification and regression tasks.</li>
</ul>
<h3 class="text-xl font-bold text-blue-700 mt-6">📌 Assumptions of Random Forest</h3>
<ul class="list-disc list-inside space-y-2">
<li>Each tree is independent and makes its own prediction.</li>
<li>Each tree is trained on random samples and features.</li>
<li>A large enough dataset is required for diverse learning.</li>
<li>Combining different trees improves accuracy.</li>
</ul>
</div>
</div>
</body>
</html>
{% endblock %} |