Interactive Geospatial Dashboard

Loading map...

Model Comparison

Class Distribution

Model Performance

U-Net

Macro-F1 93.3%

Mean IoU 87.6%. Best-performing segmentation model in the current experiment.

DeepLabV3Plus

Macro-F1 91.0%

Mean IoU 83.7%. Strong segmentation baseline, but lower than U-Net in the current experiment.

Suitability Legend

Very Low

Low

Moderate

High

Very High

Confidence Ramp

LowHigh

Selected Feature

Name: Not available

Displayed class: Not available

Rule class: Not available

Residue: Not available

Commodity: Not available

Top Residue Priority

Not available

District ranking is generated from prepared static metadata.

Prediction Confidence

Model signal

Confidence expresses model certainty and is not a substitute for field validation.

Hotspot Explorer

Rank	District	Class

Methodology and Limitations

Suitability classes and deep-learning outputs are derived from proxy labels generated from a rule-based spatial suitability index. They should be interpreted as decision-support outputs, not field-validated measurements of residue availability.

Open methodology

Loaded summary: Not available districts, Not available facilities, Not available road features.