Frequently asked questions (FAQs)

Now that we have solved the video classification problem, but with low accuracy, there are other practical aspects of this problem and overall deep learning phenomena that need to be considered too. In this section, we will see some frequently asked questions that may be on your mind. Answers to these questions can be found in Appendix A.

1. My machine has multiple GPUs installed (for example, two), but DL4J is using only one. How do I fix this problem?
2. I have configured a p2.8 xlarge EC2 GPU compute instance on AWS. However, it is showing low disk space while installing and configuring CUDA and cuDNN. How to fix this issue?
3. I understand how the distributed training happens on AWS EC2 AMI instance. However, my machine has a low-end GPU, and often I get OOP on the GPU. How can solve the issue?
4. Can I treat this application as a human activity recognition...