hive.agents.qnets.noisy_linear module

class hive.agents.qnets.noisy_linear.NoisyLinear(in_dim, out_dim, std_init=0.5)

Bases: Module

NoisyLinear Layer. Implements the layer described in https://arxiv.org/abs/1706.10295.

Parameters

• in_dim (int) – The dimension of the input.

• out_dim (int) – The desired dimension of the output.

• std_init (float) – The range for the initialization of the standard deviation of the weights.

forward(inp)

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.

training: bool