credit.losses.almost_fair_crps

credit.losses.almost_fair_crps#

Classes#

AlmostFairKCRPSLoss

Base class for all neural network modules.

Module Contents#

class credit.losses.almost_fair_crps.AlmostFairKCRPSLoss(alpha=1.0, reduction='mean', no_autocast=True)#

Bases: torch.nn.Module

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing them to be nested in a tree structure. You can assign the submodules as regular attributes:

import torch.nn as nn
import torch.nn.functional as F


class Model(nn.Module):
    def __init__(self) -> None:
        super().__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will also have their parameters converted when you call to(), etc.

Note

As per the example above, an __init__() call to the parent class must be made before assignment on the child.

Variables:

training (bool) – Boolean represents whether this module is in training or evaluation mode.

alpha = 1.0#
reduction = 'mean'#
no_autocast = True#
batched_forward#
forward(target, pred)#
single_sample_forward(target, pred)#
Parameters:

  • target – shape (1, c, t, lat, lon)

  • pred – shape (ensemble, c, t, lat, lon)

Returns:

shape (c, t, lat, lon)

Return type:

crps

_kernel_crps(preds: torch.Tensor, targets: torch.Tensor, alpha: float)#
Parameters:

  • preds – (c, t, lat, lon, ensemble)

  • targets – (c, t, lat, lon)

Returns:

(c, t, lat, lon)

Return type:

crps

Contents