aepsych.means.constant_partial_grad module

class aepsych.means.constant_partial_grad.ConstantMeanPartialObsGrad(constant_prior=None, constant_constraint=None, batch_shape=torch.Size([]), **kwargs)[source]

Bases: ConstantMean

A mean function for use with partial gradient observations.

This follows gpytorch.means.constant_mean_grad and sets the prior mean for derivative observations to 0, though unlike that function it allows for partial observation of derivatives.

The final column of input should be an index that is 0 if the observation is of f, or i if it is of df/dxi.

Initialize internal Module state, shared by both nn.Module and ScriptModule.

  • constant_prior (Optional[Prior]) –

  • constant_constraint (Optional[Interval]) –

  • batch_shape (Size) –

  • kwargs (Any) –


Define the computation performed at every call.

Should be overridden by all subclasses.


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.


input (Tensor) –

Return type:


Module contents