Corpus: TIMIT data Library: CMU Sphinx Objective: Transcribe spoken words into text with DNNs
Dataset: Breast-cancer cancer dataset dataset Library: TensorFlow
Dataset: 20 Newsgroups dataset Library: Deeplearning4j Task: Categorize documents using term representations
Dataset: AT&T face dataset Library: OpenCV library Task: Recognize face images using eigenfaces
Dataset used: CIFAR 10 dataset Toolkit: Deeplearning4j library Task description: Categorize images into 10x labels (e.g., creatures, vehicles, etc.)
Project 2: Sentiment Assessment using Recurrent Connectionist Architectures (recurrent nets) Sentiment analysis is a natural language processing (NLP) job that involves ascertaining the sentiment of a snippet of content. seq-nets are well-matched for this task, and you can use the Weka library to create an recurrent-based sentiment evaluation system.
Corpus: Equity trading data Library: Deeplearning4j Task: Predict share prices via LSTM architectures
