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