Mtzig Cross Entropy Loss

---
name: mtzig-cross-entropy-loss
description: Compute mtzig/cross_entropy_loss via the HuggingFace `evaluate` library. Use when the user has predictions + references and wants the canonical implementation of mtzig/cross_entropy_loss.
metadata:
  skill_kind: metric
  source_lib: huggingface-evaluate
  hf_module: mtzig/cross_entropy_loss
  source: library_introspection
---

# mtzig-cross-entropy-loss

> Metric `mtzig/cross_entropy_loss` from the HuggingFace `evaluate` library.

## When to invoke

User asks to compute `mtzig/cross_entropy_loss` or wants HF evaluate's canonical version.

## Recipe

```python
import evaluate
metric = evaluate.load("mtzig/cross_entropy_loss")
result = metric.compute(predictions=preds, references=refs)
print(result)
```

## Don'ts

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