FiLM2d¶

class torchelie.nn.FiLM2d(channels: int, cond_channels: int)¶

Feature-wise Linear Modulation from https://distill.pub/2018/feature-wise-transformations/ The difference with AdaIN is that FiLM does not uses the input’s mean and std in its calculations

Parameters

channels (int) – number of input channels
cond_channels (int) – number of conditioning channels from which bias and scale will be derived

condition(z: torch.Tensor) → None¶

Conditions the layer before the forward pass if z will not be present when calling forward

forward(x, z: Optional[torch.Tensor] = None) → torch.Tensor¶

Forward pass

Parameters

x (4D tensor) – input tensor
z (2D tensor, optional) – conditioning vector. If not present, condition(z) must be called first

Returns

x, conditioned