User’s Guide

• Input Data Formats and Datasets
• BigARTM Command Line Utility
• Python Tutorial
• Python Guide
  • 1. Loading Data: BatchVectorizer and Dictionary
  • 2. Base PLSA Model with Perplexity Score
  • 3. Regularizers and Scores Usage
  • 4. Multimodal Topic Models
  • 5. Phi and Theta Extraction. Transform Method
  • 6. Tokens Coocurancy and Coherence Coumputation
  • 7. Attach Model and Custom Phi Initialization
  • 8. Deal with Ptdw Matrix
  • Different Useful Techniques
• Regularizers Description
  • Smooth/Sparse Phi
  • Smooth/Sparse Theta
  • Decorrelator Phi
  • Label Regularization Phi
  • Specified sparse Phi
  • Improve Coherence Phi
  • Smooth Ptdw
  • Topic Selection Theta
  • Biterms Phi
  • Hierarchy Sparsing Theta
  • Topic Segmentation Ptdw
• Scores Description
  • Perplexity
  • Sparsity Phi
  • Sparsity Theta
  • Top Tokens
  • Topic Kernel Scores
  • Topic Mass
  • Class Presicion
  • Background Tokens Ratio
  • Items Processed (technical)
  • Theta Snippet (technical)