Once again I am here to share my take on a project. The forest cover type classification problem was an interesting one. It took me a couple of days to be clear on my approach and execution. However, here are the highlights of my contribution to the project:
- I addressed the skewed class distribution problem in the dataset using the Synthetic Minority Over-sampling Technique (SMOTE).
- I applied the Keras ModelCheckpoint callback during training to capture the best-optimized model based on validation loss.
- I handled some of the reproducibility issues inherent in using neural networks by applying the Central Limit Theorem.
Here is the link to the project: GitHub - PyYakuza/Forest-Cover-Type-Classification-NN: Implementation of Neural Networks to Forest Cover Type Discrimination
ps: please, don’t give up on your learning journey