build-with-ag2/automate-ml-for-kaggle/index.html at main · ag2ai/build-with-ag2 · GitHub

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
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>Automated Machine Learning Workflow</title>
    <script src="https://cdn.jsdelivr.net/npm/mermaid@10.6.1/dist/mermaid.min.js"></script>
    <style>
        body {
            font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
            line-height: 1.6;
            color: #333;
            max-width: 1200px;
            margin: 0 auto;
            padding: 20px;
            background: linear-gradient(135deg, #74b9ff 0%, #0984e3 50%, #6c5ce7 100%);
            min-height: 100vh;
        }

        .container {
            background: white;
            border-radius: 15px;
            box-shadow: 0 20px 40px rgba(0,0,0,0.15);
            overflow: hidden;
        }

        .header {
            background: linear-gradient(135deg, #6c5ce7 0%, #a29bfe 100%);
            color: white;
            padding: 40px;
            text-align: center;
        }

        .header h1 {
            margin: 0;
            font-size: 2.5em;
            font-weight: 300;
        }

        .header p {
            margin: 15px 0 0 0;
            opacity: 0.9;
            font-size: 1.1em;
        }

        .content {
            padding: 40px;
        }

        h2 {
            color: #2d3436;
            border-bottom: 3px solid #74b9ff;
            padding-bottom: 10px;
            margin-top: 40px;
        }

        h3 {
            color: #636e72;
            margin-top: 30px;
        }

        .tags {
            display: flex;
            flex-wrap: wrap;
            gap: 8px;
            margin: 20px 0;
        }

        .tag {
            background: #e3f2fd;
            color: #1976d2;
            padding: 5px 12px;
            border-radius: 20px;
            font-size: 0.9em;
            font-weight: 500;
        }

        .flowchart-container {
            background: #fafafa;
            border-radius: 10px;
            padding: 20px;
            margin: 30px 0;
            border: 2px solid #e0e0e0;
        }

        .mermaid {
            text-align: center;
        }

        .state-machine-grid {
            display: grid;
            grid-template-columns: repeat(auto-fit, minmax(280px, 1fr));
            gap: 20px;
            margin: 30px 0;
        }

        .state-card {
            background: linear-gradient(135deg, #f8f9fa 0%, #e9ecef 100%);
            border-left: 5px solid #6c5ce7;
            padding: 20px;
            border-radius: 8px;
            box-shadow: 0 4px 15px rgba(0,0,0,0.1);
            transition: transform 0.2s ease;
        }

        .state-card:hover {
            transform: translateY(-5px);
        }

        .state-card h4 {
            color: #5f3dc4;
            margin: 0 0 10px 0;
            font-size: 1.2em;
            display: flex;
            align-items: center;
            gap: 10px;
        }

        .state-icon {
            font-size: 1.5em;
        }

        .state-card .agents {
            background: #d1ecf1;
            padding: 15px;
            border-radius: 5px;
            margin-top: 15px;
            font-family: 'Courier New', monospace;
            font-size: 0.9em;
            border-left: 4px solid #0c5460;
        }

        .agents-label {
            font-weight: bold;
            color: #0c5460;
            margin-bottom: 8px;
        }

        .transition-info {
            background: #fff3cd;
            padding: 10px;
            border-radius: 5px;
            margin-top: 10px;
            font-size: 0.9em;
            border-left: 3px solid #ffc107;
        }

        .workflow-steps {
            background: linear-gradient(135deg, #d4edda 0%, #c3e6cb 50%);
            border-radius: 10px;
            padding: 25px;
            margin: 30px 0;
            border-left: 5px solid #28a745;
        }

        .workflow-steps h3 {
            color: #155724;
            margin-top: 0;
        }

        .workflow-step {
            background: white;
            padding: 15px;
            margin: 10px 0;
            border-radius: 8px;
            border-left: 3px solid #6c5ce7;
            box-shadow: 0 2px 5px rgba(0,0,0,0.1);
        }

        .workflow-step strong {
            color: #5f3dc4;
        }

        .code-block {
            background: #2d3748;
            color: #e2e8f0;
            padding: 20px;
            border-radius: 8px;
            overflow-x: auto;
            font-family: 'Courier New', monospace;
            margin: 20px 0;
            border-left: 4px solid #74b9ff;
        }

        .features-grid {
            display: grid;
            grid-template-columns: repeat(auto-fit, minmax(250px, 1fr));
            gap: 20px;
            margin: 30px 0;
        }

        .feature-card {
            background: linear-gradient(135deg, #e8f4f8 0%, #d4e9f7 100%);
            padding: 20px;
            border-radius: 10px;
            border-left: 4px solid #74b9ff;
        }

        .feature-card h4 {
            color: #0984e3;
            margin: 0 0 10px 0;
        }

        .links {
            background: linear-gradient(135deg, #6c5ce7 0%, #a29bfe 100%);
            color: white;
            padding: 30px;
            border-radius: 10px;
            margin: 30px 0;
        }

        .links h3 {
            color: white;
            margin-top: 0;
        }

        .links a {
            color: #e4d9ff;
            text-decoration: none;
        }

        .links a:hover {
            text-decoration: underline;
            color: #ffffff;
        }

        .links ul {
            list-style: none;
            padding: 0;
        }

        .links li {
            margin: 10px 0;
            padding-left: 20px;
            position: relative;
        }

        .links li::before {
            content: "🔗";
            position: absolute;
            left: 0;
        }

        .highlight-box {
            background: linear-gradient(135deg, #e8f4f8 0%, #d4e9f7 100%);
            border: 2px solid #74b9ff;
            border-radius: 10px;
            padding: 20px;
            margin: 20px 0;
            position: relative;
        }

        .highlight-box::before {
            content: "🎯";
            position: absolute;
            top: -10px;
            left: 20px;
            background: #74b9ff;
            color: white;
            padding: 5px 10px;
            border-radius: 5px;
            font-size: 1.2em;
        }

        .highlight-box p {
            margin: 0;
            color: #0984e3;
            font-weight: 500;
        }

        @media (max-width: 768px) {
            body {
                padding: 10px;
            }

            .header {
                padding: 30px 20px;
            }

            .header h1 {
                font-size: 2em;
            }

            .content {
                padding: 30px 20px;
            }

            .state-machine-grid, .features-grid {
                grid-template-columns: 1fr;
            }
        }
    </style>
</head>
<body>
    <div class="container">
        <div class="header">
            <h1>🤖 Automated Machine Learning Workflow</h1>
            <p>State-driven ML pipeline with intelligent agent orchestration</p>
        </div>

        <div class="content">
            <p>In this project, we build a machine learning workflow using AG2. The workflow involves data analysis, preprocessing, and model training to build a machine learning model.</p>

            <p>Machine learning workflows typically involve several key steps:</p>
            <ol>
                <li><strong>Data Analysis and Exploration</strong>: Understanding dataset size, columns, and distributions.</li>
                <li><strong>Data Preprocessing</strong>: Cleaning data, handling missing values, and encoding categorical variables.</li>
                <li><strong>Model Training</strong>: Training a model, comparing different models, and tuning hyperparameters.</li>
            </ol>

            <h2>🏗️ System Architecture</h2>

            <h3>State Machine Workflow</h3>
            <p>The system follows a state machine design with intelligent transitions between ML workflow stages:</p>

            <div class="flowchart-container">
                <div class="mermaid">
                    flowchart TD
                        A["🚀 Init State"] --> B["🔍 Explore State"]

                        B --> B1["👤 Data Explorer"]
                        B1 --> B2["⚙️ Code Executor"]
                        B2 -->|Success| C["🛠️ Preprocess State"]
                        B2 -->|Error| B1

                        C --> C1["🔧 Data Preprocessor"]
                        C1 --> C2["⚙️ Code Executor"]
                        C2 -->|LLM Decision:<br/>Ready for Training| D["🎯 Train State"]
                        C2 -->|Need More Analysis| B
                        C2 -->|Error| C1

                        D --> D1["🧠 Model Trainer"]
                        D1 --> D2["⚙️ Code Executor"]
                        D2 -->|< 2 Trials| D1
                        D2 -->|≥ 2 Trials| E["📊 Summarize State"]
                        D2 -->|Error| D1

                        E --> E1["📝 Summarizer"]
                        E1 --> F["🏁 End State"]

                        %% State grouping
                        subgraph STATES ["State Machine Workflow"]
                            direction LR
                            G["Custom Speaker<br/>Selection Method"]
                            H["StateFlow Pattern<br/>Transition Logic"]
                        end

                        %% Styling
                        classDef initEnd fill:#ffeaa7,stroke:#fdcb6e,stroke-width:3px
                        classDef state fill:#e17055,stroke:#d63031,stroke-width:3px
                        classDef agent fill:#74b9ff,stroke:#0984e3,stroke-width:2px
                        classDef executor fill:#00b894,stroke:#00a085,stroke-width:2px
                        classDef pattern fill:#fd79a8,stroke:#e84393,stroke-width:2px

                        class A,F initEnd
                        class B,C,D,E state
                        class B1,C1,D1,E1 agent
                        class B2,C2,D2 executor
                        class G,H pattern
                </div>
            </div>

            <h2>📋 State Machine Details</h2>

            <div class="state-machine-grid">
                <div class="state-card">
                    <h4><span class="state-icon">🔍</span>Explore State</h4>
                    <p>Analyze the dataset structure, distributions, and characteristics to understand the data landscape.</p>
                    <div class="agents">
                        <div class="agents-label">Agents:</div>
                        • Data Explorer → Code Executor
                    </div>
                    <div class="transition-info">
                        <strong>Transition:</strong> Success → Preprocess | Error → Stay in Explore
                    </div>
                </div>

                <div class="state-card">
                    <h4><span class="state-icon">🛠️</span>Preprocess State</h4>
                    <p>Clean and prepare data including handling missing values, encoding categoricals, and feature scaling.</p>
                    <div class="agents">
                        <div class="agents-label">Agents:</div>
                        • Data Preprocessor → Code Executor
                    </div>
                    <div class="transition-info">
                        <strong>Transition:</strong> LLM decides if ready for training → Train | Need more analysis → Explore
                    </div>
                </div>

                <div class="state-card">
                    <h4><span class="state-icon">🎯</span>Train State</h4>
                    <p>Train and compare multiple ML models with different algorithms and hyperparameters.</p>
                    <div class="agents">
                        <div class="agents-label">Agents:</div>
                        • Model Trainer → Code Executor
                    </div>
                    <div class="transition-info">
                        <strong>Transition:</strong> < 2 trials → Continue training | ≥ 2 trials → Summarize | Error → Retry
                    </div>
                </div>

                <div class="state-card">
                    <h4><span class="state-icon">📊</span>Summarize State</h4>
                    <p>Generate comprehensive workflow summary and integrate all successful code snippets.</p>
                    <div class="agents">
                        <div class="agents-label">Agents:</div>
                        • Summarizer (LLM-only)
                    </div>
                    <div class="transition-info">
                        <strong>Transition:</strong> Always → End
                    </div>
                </div>
            </div>

            <div class="workflow-steps">
                <h3>🔄 Workflow Process</h3>
                <div class="workflow-step">
                    <strong>Step 1 - Dataset Analysis:</strong> Explore data shape, types, distributions, and missing values
                </div>
                <div class="workflow-step">
                    <strong>Step 2 - Data Preprocessing:</strong> Handle missing data, encode categoricals, and scale features
                </div>
                <div class="workflow-step">
                    <strong>Step 3 - Model Training:</strong> Train multiple models (2 iterations) with performance comparison
                </div>
                <div class="workflow-step">
                    <strong>Step 4 - Visualization:</strong> Generate performance plots, confusion matrices, and evaluation metrics
                </div>
                <div class="workflow-step">
                    <strong>Step 5 - Code Integration:</strong> Combine all successful code into reproducible script
                </div>
            </div>

            <h2>🤖 AG2 Features</h2>
            <div class="features-grid">
                <div class="feature-card">
                    <h4>🎭 Custom Speaker Transitions</h4>
                    <p>State-driven agent selection using custom speaker_selection_method for workflow control</p>
                </div>
                <div class="feature-card">
                    <h4>🌊 StateFlow Design</h4>
                    <p>Build state-driven workflows with intelligent transitions based on execution results</p>
                </div>
                <div class="feature-card">
                    <h4>⚡ Code Execution</h4>
                    <p>Jupyter-based code execution environment for interactive ML development</p>
                </div>
                <div class="feature-card">
                    <h4>🧠 LLM Decision Making</h4>
                    <p>AI-powered decisions on workflow readiness and state transitions</p>
                </div>
            </div>

            <h2>🏷️ Tags</h2>
            <div class="tags">
                <span class="tag">data-analysis</span>
                <span class="tag">groupchat</span>
                <span class="tag">stateflow</span>
                <span class="tag">code-execution</span>
                <span class="tag">kaggle</span>
                <span class="tag">automated-ml</span>
                <span class="tag">workflow-automation</span>
                <span class="tag">model-training</span>
                <span class="tag">data-preprocessing</span>
                <span class="tag">state-machine</span>
                <span class="tag">hyperparameter-tuning</span>
            </div>

            <h2>📋 Prerequisites</h2>
            <ul>
                <li>Python 3.12 or higher</li>
                <li>OpenAI API key</li>
            </ul>

            <h2>⚙️ Installation</h2>
            <ol>
                <li>Clone and navigate to the folder:</li>
            </ol>
            <div class="code-block">git clone https://github.com/ag2ai/build-with-ag2.git
cd build-with-ag2/automate-ml-for-kaggle</div>

            <ol start="2">
                <li>Install dependencies:</li>
            </ol>
            <div class="code-block">uv sync</div>

            <ol start="3">
                <li>Set up environment variables:</li>
            </ol>
            <div class="code-block">cp .env.example .env
# Edit .env with your OpenAI API key</div>

            <h2>🚀 Usage</h2>
            <div class="code-block">uv run python main.py</div>

            <div class="highlight-box">
                <p><strong>Automated Workflow:</strong> The system will automatically analyze the dataset (house_prices_train.csv), preprocess the data, train and compare multiple models, generate performance visualizations, and output a comprehensive summary.</p>
            </div>

            <p>The workflow will:</p>
            <ol>
                <li>Analyze the dataset (<code>house_prices_train.csv</code>)</li>
                <li>Preprocess the data automatically</li>
                <li>Train and compare multiple models</li>
                <li>Generate performance visualizations</li>
                <li>Output a comprehensive summary</li>
            </ol>

            <div class="links">
                <h3>📞 Contact</h3>
                <p>For more information or any questions, please refer to the documentation or reach out to us!</p>
                <ul>
                    <li><a href="https://docs.ag2.ai/latest/">View Documentation at: https://docs.ag2.ai/latest/</a></li>
                    <li><a href="https://github.com/ag2ai/ag2">Find AG2 on github: https://github.com/ag2ai/ag2</a></li>
                    <li><a href="https://discord.gg/pAbnFJrkgZ">Join us on Discord: https://discord.gg/pAbnFJrkgZ</a></li>
                    <li><a href="mailto:support@ag2.ai">Email us at: support@ag2.ai</a></li>
                </ul>
            </div>

            <h2>📄 License</h2>
            <p>This project is licensed under the Apache License 2.0. See the <a href="../LICENSE">LICENSE</a> for details.</p>
        </div>
    </div>

    <script>
        mermaid.initialize({
            startOnLoad: true,
            theme: 'default',
            flowchart: {
                useMaxWidth: true,
                htmlLabels: true,
                curve: 'basis'
            }
        });
    </script>
</body>
</html>