top5_error_rate.py

from typing import Dict, Any

import datasets
import evaluate
import numpy as np
from evaluate.utils.file_utils import add_start_docstrings

_DESCRIPTION = """
The "top-5 error" is the percentage of times that the target label does not appear among the 5 highest-probability predictions. It can be computed with:
Top-5 Error Rate = 1 - Top-5 Accuracy
or equivalently:
Top-5 Error Rate = (Number of incorrect top-5 predictions) / (Total number of cases processed)
 Where:
- Top-5 Accuracy: The proportion of cases where the true label is among the model's top 5 predicted classes.
- Incorrect top-5 prediction: The true label is not in the top 5 predicted classes (ranked by probability).
"""

_KWARGS_DESCRIPTION = """
Args:
    predictions (`list` of `list` of `int`): Predicted labels. Each inner list should contain the top-5 predicted class indices.
    references (`list` of `int`): Ground truth labels.
Returns:
    top5_error_rate (`float`): Top-5 Error Rate score. Minimum possible value is 0. Maximum possible value is 1.0.
Examples:
    >>> metric = evaluate.load("top5_error_rate")
    >>> results = metric.compute(
    ...     references=[0, 1, 2], 
    ...     predictions=[[0, 1, 2, 3, 4], [1, 0, 2, 3, 4], [2, 0, 1, 3, 4]]
    ... )
    >>> print(results)
    {'top5_error_rate': 0.0}
"""

_CITATION = """
"""


@add_start_docstrings(_DESCRIPTION, _KWARGS_DESCRIPTION)
class Top5ErrorRate(evaluate.Metric):
    def _info(self):
        return evaluate.MetricInfo(
            description=_DESCRIPTION,
            citation=_CITATION,
            inputs_description=_KWARGS_DESCRIPTION,
            features=datasets.Features(
                {
                    "predictions": datasets.Sequence(datasets.Sequence(datasets.Value("float32"))),
                    "references": datasets.Sequence(datasets.Value("int32")),
                }
                if self.config_name == "multilabel"
                else {
                    "predictions": datasets.Sequence(datasets.Value("float32")),
                    "references": datasets.Value("int32"),
                }
            ),
            reference_urls=[],
        )

    def _compute(
            self,
            *,
            predictions: list[list[float]] = None,
            references: list[int] = None,
            **kwargs,
    ) -> Dict[str, Any]:
        # to numpy array
        outputs = np.array(predictions, dtype=np.float32)
        labels = np.array(references)

        # Top-1 ACC
        pred = outputs.argmax(axis=1)
        acc = (pred == labels).mean()

        # Top-5 Error Rate
        top5_indices = outputs.argsort(axis=1)[:, -5:]
        correct = (labels.reshape(-1, 1) == top5_indices).any(axis=1)
        top5_error_rate = 1 - correct.mean()

        return {
            "accuracy": float(acc),
            "top5_error_rate": float(top5_error_rate)
        }