5.5. clumsygrad.loss#

This module provides functions to compute various loss functions.

clumsygrad.loss.mse_loss(pred: Tensor, target: Tensor) → Tensor[source]#

Computes the Mean Squared Error (MSE) loss between the predicted and target tensors.

Parameters:

  • pred – The predicted tensor.

  • target – The target tensor.

Returns:

The MSE loss tensor.

Return type:

Tensor

clumsygrad.loss.mae_loss(pred: Tensor, target: Tensor) → Tensor[source]#

Computes the Mean Absolute Error (MAE) loss between the predicted and target tensors.

Parameters:

  • pred – The predicted tensor.

  • target – The target tensor.

Returns:

The MAE loss tensor.

Return type:

Tensor