Source code for catalyst.tools.meters.confusionmeter

"""
Maintains a confusion matrix for a given classification problem.
"""
import numpy as np

import torch

from catalyst.tools.meters import meter


class ConfusionMeter(meter.Meter):
    """
    ConfusionMeter constructs a confusion matrix for a multiclass
    classification problems. It does not support multilabel, multiclass
    problems: for such problems, please use MultiLabelConfusionMeter.
    """

    def __init__(self, k: int, normalized: bool = False):
        """
        Args:
            k: number of classes in the classification problem
            normalized: determines whether or not the confusion
                matrix is normalized or not
        """
        super(ConfusionMeter, self).__init__()
        self.conf = np.ndarray((k, k), dtype=np.int32)
        self.normalized = normalized
        self.k = k
        self.reset()

    def reset(self) -> None:
        """Reset confusion matrix, filling it with zeros."""
        self.conf.fill(0)

    def add(self, predicted: torch.Tensor, target: torch.Tensor) -> None:
        """Computes the confusion matrix of ``K x K`` size
        where ``K`` is no of classes.

        Args:
            predicted: Can be an N x K tensor of predicted scores
                obtained from the model for N examples and K classes
                or an N-tensor of integer values between 0 and K-1
            target: Can be a N-tensor of integer values assumed
                to be integer values between 0 and K-1 or N x K tensor, where
                targets are assumed to be provided as one-hot vectors
        """
        predicted = predicted.cpu().numpy()
        target = target.cpu().numpy()

        assert (
            predicted.shape[0] == target.shape[0]
        ), "number of targets and predicted outputs do not match"

        if np.ndim(predicted) != 1:
            assert (
                predicted.shape[1] == self.k
            ), "number of predictions does not match size of confusion matrix"
            predicted = np.argmax(predicted, 1)
        else:
            assert (predicted.max() < self.k) and (
                predicted.min() >= 0
            ), "predicted values are not between 1 and k"

        onehot_target = np.ndim(target) != 1
        if onehot_target:
            assert (
                target.shape[1] == self.k
            ), "Onehot target does not match size of confusion matrix"
            assert (target >= 0).all() and (
                target <= 1
            ).all(), "in one-hot encoding, target values should be 0 or 1"
            assert (
                target.sum(1) == 1
            ).all(), "multilabel setting is not supported"
            target = np.argmax(target, 1)
        else:
            assert (predicted.max() < self.k) and (
                predicted.min() >= 0
            ), "predicted values are not between 0 and k-1"

        # hack for bincounting 2 arrays together
        x = predicted + self.k * target
        bincount_2d = np.bincount(  # noqa: WPS114
            x.astype(np.int32), minlength=self.k ** 2
        )
        assert bincount_2d.size == self.k ** 2
        conf = bincount_2d.reshape((self.k, self.k))

        self.conf += conf

    def value(self):
        """
        Returns:
            Confusion matrix of K rows and K columns, where rows corresponds
            to ground-truth targets and columns corresponds to predicted
            targets.
        """
        if self.normalized:
            conf = self.conf.astype(np.float32)
            return conf / conf.sum(1).clip(min=1e-12)[:, None]
        else:
            return self.conf


__all__ = ["ConfusionMeter"]