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