qhjqhj00 lvwerra-test
Metric
lvwerra/testfrom the HuggingFaceevaluatelibrary.
When to invoke
User asks to compute lvwerra/test or wants HF evaluate's canonical version.
Recipe
import evaluate
metric = evaluate.load("lvwerra/test")
result = metric.compute(predictions=preds, references=refs)
print(result)
Don'ts
- Don't assume your in-house
lvwerra/testmatches HF — version conventions vary. - Many evaluate metrics have task-specific arguments (
average=,lang=,model_type=); read the metric card before reporting numbers.