Sea Ai Panoptic Quality

---
name: sea-ai-panoptic-quality
description: Compute SEA-AI/panoptic-quality via the HuggingFace `evaluate` library. Use when the user has predictions + references and wants the canonical implementation of SEA-AI/panoptic-quality.
metadata:
  skill_kind: metric
  source_lib: huggingface-evaluate
  hf_module: SEA-AI/panoptic-quality
  source: library_introspection
---

# sea-ai-panoptic-quality

> Metric `SEA-AI/panoptic-quality` from the HuggingFace `evaluate` library.

## When to invoke

User asks to compute `SEA-AI/panoptic-quality` or wants HF evaluate's canonical version.

## Recipe

```python
import evaluate
metric = evaluate.load("SEA-AI/panoptic-quality")
result = metric.compute(predictions=preds, references=refs)
print(result)
```

## Don'ts

- Don't assume your in-house `SEA-AI/panoptic-quality` matches HF — version conventions vary.
- Many evaluate metrics have task-specific arguments (`average=`, `lang=`, `model_type=`); read the metric card before reporting numbers.