wrappers

class ledidi.wrappers.DesignWrapper(models)

A wrapper for using multiple models in design.

This wrapper will accept multiple models and turn their predictions into a vector. A requirement is that the output from each individual model is a tensor whose last dimension is 1, and that all of the models have the same other dimensions. For instance, if three models are passed in that each make predictions of shape (batch_size, 1), the return from this wrapper would have shape (batch_size, 3).

This wrapper is used to design edits when you want to balance the predictions from several models, e.g., by increasing predictions from one model without changing predictions from a second model. In practice, one would now pass in a vector of desired targets instead of a single value and the loss would be calculated over each of them.

Parameters

models: list, tuple

A set of torch.nn.Module objects.

forward(X)

Define 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.