Pinned Loading

1. dog_app dog_app Public

   Convolutional Neural Networks (CNN) are designed via transfer learning using bottleneck features from VGG-16 and ResNet-50 architectures. An accuracy of around 82% is achieved for the prediction of…

   Jupyter Notebook 1

2. smartcab smartcab Public

   In this Reinforcement Learning (RL) project, an epsilon-greedy Q-Learning agent is constructed to learn an optimal policy for smartcabs.

   Jupyter Notebook

3. customer_segments customer_segments Public

   The Principal Component Analysis (PCA) is implemented on a dataset (from 440 clients of a wholesale distributor with variables related to its clients' total spending values on different product cat…

   Jupyter Notebook 1

4. finding_donors finding_donors Public

   Different classification algorithms are used to predict income levels of potential donors. The hyperparameters of the classifiers are optimized via grid_search and, among others, an ensemble classi…

   HTML 1

5. boston_housing boston_housing Public

   An exploratory data analysis is performed and a regression model is used to predict house values. The prediction performance is optimized after tuning the model hyper-parameters to minimize bias/va…

   HTML

6. spam-email spam-email Public

   Features are extracted from text data using the bag of words model and a Naive Bayes classifier is trained and used to predict email spams.

   Jupyter Notebook 1