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