Projects for Neural Networks class ECE NTUA 2021-2022.
-
1st project (Supervised learning):
We used two datasets (UCI and Kaggle) and experimented with various classifiers (dummy, k-nearest neighbors, gaussian naive bayes, Logistic Regression) in order to solve a classification problem. We also tuned the hyperparameters of each model using grid search cross validation. -
2nd project (Unsupervised learning):
We implemented a movie recommendation systme using word emmbeddings and then implemented topological and semantic movie illustration using self organizing maps (S0M). -
3rd project (Deep learning):
We implemented an image captioning system. We tuned the hyperparameters of the CNN models (InceptionV3 and NASNetLarge) and added more layers when it was deemed necessary.