Video model

class expert.core.congruence.video_emotions.video_model.DAN(num_class: int = 8, num_head: int = 4, pretrained: bool = True, device: torch.device | None = None)

Bases: torch.nn.modules.module.Module

Distract Your Attention Network implementation.

Distract Your Attention Network performs facial expression recognition on tensor face images.

property device: torch.device

Check the device type.

Returns

Device type on local machine.

Return type

torch.device

forward(x: torch.Tensor) → Tuple[torch.Tensor, torch.Tensor, torch.Tensor]

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class expert.core.congruence.video_emotions.video_model.CrossAttentionHead

Bases: torch.nn.modules.module.Module

forward(x: torch.Tensor) → torch.Tensor

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class expert.core.congruence.video_emotions.video_model.SpatialAttention

Bases: torch.nn.modules.module.Module

forward(x: torch.Tensor) → torch.Tensor

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class expert.core.congruence.video_emotions.video_model.ChannelAttention

Bases: torch.nn.modules.module.Module

forward(sa: torch.Tensor) → torch.Tensor

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.