This hands-on course introduces data scientists to technologies related to building and operating live, high throughput deep learning applications running on powerful servers in the cloud as well on smaller and lower power devices at the edge of the network. The material of the class is a set of practical approaches, code recipes, and lessons learned. It is based on the latest developments in the industry and industry use cases as opposed to pure theory.
See hw01 for edge device requirements.
The syllabus and homeworks are as follows,
| Week | Content |
|---|---|
| 01 | Introduction and Overview |
| 02 | Clouds, Infrastructure, and Machine Learning Cloud Services |
| 03 | Introduction to Containers |
| 04 | Deep Learning 101 |
| 05 | Deep Learning Frameworks |
| 06 | Optimizing Models for the Edge and GStreamer |
| 07 | Deep Learning 201 |
| 08 | Datasets and Dataset Processing |
| 09 | HPC, MPI, and Multinode/MultiGPU (MNMG) Training |
| 10 | Generative Adversarial Networks (GANs) |
| 11 | Deep Reinforcement Learning |
| 12 | Speech, Natural Language Processing, and Conversational Design |
| 13 | AI and DL: Applying AI to Real World Applications |