Java Deep Learning Projects

Distributed deep learning – video classification using a convolutional-LSTM network

In this project, we developed a complete deep learning application that classifies a large collection of a video dataset from the UCF101 dataset. We applied a combined CNN-LSTM network with deeplearning4j (DL4J) that overcame the limitations of standalone CNN or RNN Long Short-Term Memory (LSTM) networks.

Finally, we saw how to perform training in both a parallel and distributed manner across multiple devices (CPUs and GPUs) on AWS EC2 AMI 9.0. We performed parallel and distributed training on a p2.8xlarge instance with 8 GPUs, 32 computing cores, and 488 GB of RAM. 

One of the greatest takeaways from this chapter was that this end-to-end project can be treated ...