The latest articles on DEV Community by Ankit Kumar (@ankit02327). https://dev.to/ankit02327 https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4143553%2F1a7a7386-df67-45a9-817e-2032186c81ae.jpg https://dev.to/ankit02327 en Ankit Kumar Fri, 25 Sep 2026 21:38:36 +0000 https://dev.to/ankit02327/i-built-a-stock-price-prediction-app-while-learning-ml-heres-what-i-learned-19ho https://dev.to/ankit02327/i-built-a-stock-price-prediction-app-while-learning-ml-heres-what-i-learned-19ho <p><strong>TL;DR:</strong> I built a full-stack stock prediction app with 7 ML algorithms, 1,001 stocks, and offline-first support. It's open source, and I'd love your feedback.</p> <p>🔗 <strong><a href="https://github.com/ankit02327/stock-price" rel="noopener noreferrer">github.com/ankit02327/stock-price</a></strong></p> <p><a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fojz8ser1o8kkz2y1g99w.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Farticles%2Fojz8ser1o8kkz2y1g99w.png" alt=" " width="800" height="1016"></a></p> <h2> How It Started </h2> <p>This project began from something I was personally learning: machine learning.</p> <p>While taking ML courses at university, I wanted to understand how the algorithms I was studying could be combined into a complete, practical application. That led me to build <strong>Stock Price Prediction</strong> around a real-world problem instead of keeping the work inside individual coursework exercises.</p> <p>The educational purpose has remained central. I want someone learning machine learning to be able to look at a real application, understand how different models are used, work with real financial data, experiment with the code, and eventually build projects of their own.</p> <h2> What It Actually Does </h2> <ul> <li> <strong>7 ML algorithms:</strong> Linear Regression, Decision Tree, Random Forest, SVM (basic models) + KNN, ARIMA, Autoencoder (advanced models)</li> <li> <strong>1,001 stocks:</strong> 501 US + 500 Indian (936 used for training after filtering insufficient data)</li> <li> <strong>Real-time data:</strong> US stocks via Finnhub, Indian stocks via Upstox, with permanent offline storage fallback</li> <li> <strong>5-year historical analysis:</strong> Interactive charts with Recharts</li> <li> <strong>38 technical indicators</strong> calculated from historical data</li> <li> <strong>Currency conversion:</strong> Automatic USD/INR conversion</li> <li> <strong>Smart training:</strong> Percentage-based predictions with confidence scoring</li> </ul> <h2> The Tech Stack </h2> <div class="table-wrapper-paragraph"><table> <thead> <tr> <th>Layer</th> <th>Technology</th> </tr> </thead> <tbody> <tr> <td>Backend</td> <td>Flask 2.3.3, Python 3.8+, TensorFlow 2.20, scikit-learn, statsmodels</td> </tr> <tr> <td>Frontend</td> <td>React 18, TypeScript, Vite, Tailwind CSS, Recharts</td> </tr> <tr> <td>APIs</td> <td>Finnhub (US), Upstox (India), yfinance (historical)</td> </tr> </tbody> </table></div> <h2> What Makes It Different: Offline-First </h2> <p>The system works <strong>completely offline without any API keys</strong>. This was a deliberate design choice — I wanted anyone to be able to clone the repo and start experimenting immediately, without signing up for API keys or worrying about rate limits.</p> <p><strong>What works offline:</strong></p> <ul> <li>Stock info for 1,001 stocks</li> <li>Complete 5-year historical charts (2020–2024)</li> <li>All trained ML models</li> <li>Full-text search</li> <li>38 technical indicators</li> </ul> <p><strong>What needs API keys:</strong> Live prices only.</p> <h2> What I Learned </h2> <ol> <li><p><strong>Machine learning is not magic.</strong> It finds patterns, but you need domain knowledge to make sense of them.</p></li> <li><p><strong>Data quality is everything.</strong> Stock data is messy — splits, dividends, missing values. A robust preprocessing pipeline matters more than a fancy model.</p></li> <li><p><strong>Start simple, then iterate.</strong> My first model was basic linear regression. It wasn't great, but it worked, and it gave me a foundation to build on.</p></li> <li><p><strong>Open source is a two-way street.</strong> The project now has <strong>27 contributors</strong> and <strong>14 releases</strong>. I've learned as much from the community as from building it.</p></li> </ol> <h2> What's Next </h2> <p>I'm planning substantial improvements: FastAPI migration, Redis caching, PostgreSQL, Docker Compose, MLflow experiment tracking, news sentiment analysis, and a backtesting system.</p> <p>The project is still growing. I see its current US and Indian market coverage as a starting point rather than the final scope.</p> <h2> How You Can Help </h2> <p>If this sounds interesting:</p> <ul> <li>⭐ <strong>Star the repo:</strong> <a href="https://github.com/ankit02327/stock-price" rel="noopener noreferrer">github.com/ankit02327/stock-price</a> </li> <li>🐛 <strong>Open an issue</strong> if you find a bug</li> <li>🔧 <strong>Contribute</strong> — there are "good first issue" labels for newcomers</li> <li>📣 <strong>Share it</strong> with anyone learning ML</li> </ul> <h2> Let's Connect </h2> <ul> <li> <strong>GitHub:</strong> <a href="https://github.com/ankit02327" rel="noopener noreferrer">@ankit02327</a> </li> </ul> <p><em>Thanks for reading. If you're learning ML too, I'd love to hear what you're building.</em></p> learning machinelearning opensource showdev