Source: google-ml-course

Data in ML

• Sampling data
• Batching data
• Data representation
  • One-hot encoding
  • Scaling features
  • Binning
  • Feature crossing

Choosing labels and features

Use quantifiable, observable data.