Architecture Reference¶

Architecture¶

class dacapo.experiments.architectures.Architecture(*args, **kwargs)¶

property eval_shape_increase: Coordinate¶: How much to increase the input shape during prediction.

abstract property input_shape: Coordinate¶: The spatial input shape (i.e., not accounting for channels and batch dimensions) of this architecture.

abstract property num_in_channels: int¶: Return the number of input channels this architecture expects.

abstract property num_out_channels: int¶: Return the number of output channels of this architecture.

class dacapo.experiments.architectures.CNNectomeUNet(architecture_config)¶

property eval_shape_increase¶: How much to increase the input shape during prediction.

forward(x)¶

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.

property input_shape¶: The spatial input shape (i.e., not accounting for channels and batch dimensions) of this architecture.

property num_in_channels: int¶: Return the number of input channels this architecture expects.

property num_out_channels: int¶: Return the number of output channels of this architecture.