This project demonstrates how a neural network (MLPClassifier from 'scikit-learn') learns to classify non-linearly separable data using the 'make_circles' dataset. An interactive slider powered by 'ipywidgets' allows real-time adjustment of the hidden layer size, helping visualize changes in the decision boundary.
- Understanding neural networks through visualization
- How hidden layer size affects decision boundaries
- Use of 'make_circles' for non-linear classification examples
- Interactive widgets for making learning more engaging
- Python (scikit-learn, matplotlib, numpy)
- Google Colab
- 'ipywidgets' for interactivity
This work was completed as part of my learning journey with Oracle University, where I’ve been exploring foundational concepts in machine learning and neural networks.
- Open the notebook in Google Colab or Jupyter Notebook.
- Run all cells and use the slider to adjust the hidden layer size.
- Watch the decision boundary change dynamically!
Feel free to contribute or fork the project.