ray.rllib.utils.torch_utils.explained_variance#

ray.rllib.utils.torch_utils.explained_variance(y: numpy.array | jnp.ndarray | tf.Tensor | torch.Tensor, pred: numpy.array | jnp.ndarray | tf.Tensor | torch.Tensor) → numpy.array | jnp.ndarray | tf.Tensor | torch.Tensor[source]#

Computes the explained variance for a pair of labels and predictions.

The formula used is: max(-1.0, 1.0 - (std(y - pred)^2 / std(y)^2))

Parameters:

  • y – The labels.

  • pred – The predictions.

Returns:

The explained variance given a pair of labels and predictions.