Compute murinj/hter via the HuggingFace `evaluate` library. Use when the user has predictions + references and wants the canonical implementation of murinj/hter.
Installs into .claude/skills of the current project.
Are you the author of Murinj Hter?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/qhjqhj00-murinj-hter)
---
name: murinj-hter
description: Compute murinj/hter via the HuggingFace `evaluate` library. Use when the user has predictions + references and wants the canonical implementation of murinj/hter.
metadata:
skill_kind: metric
source_lib: huggingface-evaluate
hf_module: murinj/hter
source: library_introspection
---
# murinj-hter
> Metric `murinj/hter` from the HuggingFace `evaluate` library.
## When to invoke
User asks to compute `murinj/hter` or wants HF evaluate's canonical version.
## Recipe
```python
import evaluate
metric = evaluate.load("murinj/hter")
result = metric.compute(predictions=preds, references=refs)
print(result)
```
## Don'ts
- Don't assume your in-house `murinj/hter` matches HF — version conventions vary.
- Many evaluate metrics have task-specific arguments (`average=`, `lang=`, `model_type=`); read the metric card before reporting numbers.
