- Kaggle competition: https://www.kaggle.com/c/ift3395-6390-f2018
- Download the data by clicking on the Data tab in the link and place it in the Data folder
- Convnet logic based on this series https://www.youtube.com/watch?v=ViO56ASqeks
Link to the report: https://docs.google.com/document/d/1f1StX7hkJi4Y4KP9-WWH5prHs4W0INmgQcy9iTXqC48/edit#
Before you begin, you should familiarize yourself with the dataset.
- Data pre-processing: get rid of the noise
- Implement the classifiers naively without any optimisation/strategy
- baseline linear learner: SVM or logistic regression
- other method: Random forest and convnet
- Watch the results
- Optimization: Try to optimize Random forest and Convnet: Make decisions about training/validation split, distribution choice for naive bayes, regularization strategy, any optimization tricks, setting hyper-parameters, etc.
- Watch the results
- And so on..
- Take the one that gives the best results between Random forest and Convnet
- Compare with the baseline linear learner