Full Stack Deep Learning

1. Why do so many projects fail?

• ML is still research - you shouldn’t aim for 100% success rate
• Technically infeasible or poorly scoped
• Never make the leap to production
• Unclear success criteria
• Poor team management

[Module overview]

2. Lifecycle

Planning & Project Setup

• Decide to work on pose estimation
• Determine requirements & goals
• Allocate resources
• Consider the ethical implications
• Etc.

Data Collection & Labeling

• Collect training objects
• Set up sensors