Manuals

• 1. Datasets
  • 1.1. Reviews Datasets
  • 1.2. Twitter Sentiment Datasets
  • 1.3. UCI Machine Learning
  • 1.4. LeQua 2022 Datasets
  • 1.5. IFCB Plankton dataset
  • 1.6. Adding Custom Datasets
• 2. Evaluation
  • 2.1. Error Measures
  • 2.2. Evaluation Protocols
• 3. Explicit Loss Minimization
• 4. Quantification Methods
  • 4.1. Aggregative Methods
  • 4.2. Composable Methods
  • 4.3. Meta Models
• 5. Model Selection
  • 5.1. Targeting a Quantification-oriented loss
  • 5.2. Targeting a Classification-oriented loss
• 6. Plotting
  • 6.1. Diagonal Plot
  • 6.2. Quantification bias
  • 6.3. Error by Drift
• 7. Protocols
  • 7.1. Artificial-Prevalence Protocol
  • 7.2. Sampling from the unit-simplex, the Uniform-Prevalence Protocol (UPP)
  • 7.3. Natural-Prevalence Protocol
  • 7.4. Other protocols