Python implementation of a L layered Deep Neural Network from scratch for a binary classification problem from scratch.
Testing accuracy of 80% achieved with a 4 layer model on Cats vs Non cat image dataset. Dataset is linked in the notebook. This notebook is inspired by Andrew NG's course Neural Networks and Deep Learning and contains the same nomenclature for the equation variables.
Steps Demonstrated:
- Input setup
- Parameter Initialization
- Forward Propagation
- Backward Propagation
- Parameter Updation
- Prediction using learned parameters