LSTM

O projeto consiste em dois scripts que servem como base para complementar o fluxo da aplicação como um todo. Ambos os scripts são utilizados para facilitar e acelerar o processo de criação de quizzes e a ordenação das próximas palavras na revisão de um flashcard. São utilizadas bibliotecas como gensim e keras para a implementação dos scripts

Application of Deep Learning methods to Time Series

Build a Bitcoin predictive model. (Inputs : Klines, Google Trend data. Future inputs implementation : Liquidations data)

TensorFlow examples for Artificial Neural Networks course including: MLP with softmax output layer; MLP and CNN for MNIST dataset; CNN for CIFAR-10 dataset with data augmentation; LSTM with CNN layer for IMDB sentiment classification task.

Build a seq2seq model (with/without attention) to learn normal trajectory of ROS-based robot, then use the trained model for detection with real-time input data. This project is going along side with other ros-based node for simulating cyber-attack.

A natural language processing implementation on movie genre classification by synopsis.

Time series classification to identify relapse / non-relapse after cessation of leukemia treatment

The final project for the Data Science Lab course @DataScience Bicocca