Built a neural network and used it to predict daily bike rental ridership.

Classified images from the CIFAR-10 dataset. The dataset consists of airplanes, dogs, cats, and other objects. Preprocessed the images, then train a convolutional neural network on all the samples. The images were normalized and the labels one-hot encodeded. Built a convolutional neural network with max pooling, dropout, and fully connected layers.

Generating TV scripts using Recurrent Neural Networks

Trained a sequence to sequence model on a dataset of English and French sentences that can translate new sentences from English to French.

Used generative adversarial networks to generate new images of faces.