Issues

• Connect the application to Supabase and deploy a fully functional live demonstration

  ShreeGattani/NOAH#16 · Filter by author sarthak-jain37 opened on Sep 27, 2026

• processing.py bypasses the anomaly-fusion and classification pipeline

  ShreeGattani/NOAH#15 · Filter by author sarthak-jain37 opened on Sep 27, 2026

• Redesign Database Schema to Fully Store Anomaly Intelligence

  ShreeGattani/NOAH#14 · Filter by author sarthak-jain37 opened on Sep 27, 2026

• Final classification can lose useful evidence about simultaneous anomalies

  ShreeGattani/NOAH#13 · Filter by author sarthak-jain37 opened on Sep 27, 2026

• ML anomaly score is relative to the training distribution and may require calibration

  ShreeGattani/NOAH#12 · Filter by author sarthak-jain37 opened on Sep 27, 2026

• Missing evidence should not automatically behave like zero evidence during weighted fusion

  ShreeGattani/NOAH#11 · Filter by author sarthak-jain37 opened on Sep 27, 2026

• Cross-sensor evidence currently duplicates the same dew-point consistency evidence for temperature and humidity

  ShreeGattani/NOAH#10 · Filter by author sarthak-jain37 opened on Sep 27, 2026

• Spatial evidence is aggregated across variables instead of preserving variable-level anomaly information

  ShreeGattani/NOAH#9 · Filter by author sarthak-jain37 opened on Sep 27, 2026

• Temporal frozen-sensor detection is not sufficiently represented in final anomaly fusion

  ShreeGattani/NOAH#8 · Filter by author sarthak-jain37 opened on Sep 27, 2026

• Improve Cross-Sensor Consistency Scoring and Evidence Handling

  ShreeGattani/NOAH#7 · Filter by author sarthak-jain37 opened on Sep 27, 2026

• Improve drift detection

  ShreeGattani/NOAH#6 · Filter by author sarthak-jain37 opened on Sep 8, 2026

• Improve spike detection logic and handle low-variance windows

  ShreeGattani/NOAH#5 · Filter by author sarthak-jain37 opened on Sep 8, 2026