ML & DL Explorer is an interactive Streamlit web application that lets you train, tune, and evaluate machine learning and deep learning models — directly in your browser, with no coding required.
Upload your own dataset or use one of the built-in examples. Select a model, adjust its hyperparameters via intuitive sliders and dropdowns, and instantly visualize performance metrics and training results. The app supports both classification and regression tasks.
- Upload your own dataset or select from built-in ready-to-use datasets
- Train multiple models with a single click — no code required
- Tune hyperparameters interactively via the sidebar UI
- Visualize results — training curves, confusion matrices, correlation heatmaps, and more
- Supports both classification and regression tasks
- Fully runs in the browser via Streamlit
| Model | Task |
|---|---|
| Deep Neural Network | Classification / Regression |
| Decision Tree | Classification / Regression |
| Linear Regression | Regression |
| Logistic Regression | Classification |
- Select a task — choose Classification or Regression from the sidebar
- Load data — upload a CSV file or pick a built-in dataset
- Choose a model — select from the available ML / DL algorithms
- Tune hyperparameters — adjust settings interactively in the sidebar
- Train & evaluate — hit Run and explore the results and visualizations
Note: The entire application currently lives in a single file. Modularization into separate components (data loading, model definitions, visualization) is planned for a future update.
The app is not currently deployed. To run it yourself:
- Locally — follow the Getting Started steps above
- Streamlit Cloud — fork the repo, go to share.streamlit.io, connect your GitHub, and deploy
streamlit_ML_app.pywith one click — it's free
All code is in one file so i need to modulate code.
machine-learning deep-learning streamlit web-application
classification regression decision-trees linear-regression
logistic-regression deep-neural-networks heatmap interactive
This project is licensed under the MIT License. See LICENSE for details.