Register hooks to the trainer. The hooks are executed in the order they are registered. Args: hooks (list[Option

resize position embedding with class token example: 224:(14x14+1)-> 384: (24x24+1) return: new position embedding

Create configs and perform basic setups.

Load custom environment setup by importing a Python source file or a module, and run the setup function.

Update the model with precise statistics. Users can manually call this method.

Upgrade a config from its current version to a newer version. Args: cfg (CfgNode): to_version (int): defaults to the latest

add an instance (e.g., from another sentence).

:param d_model: Output dimensionality of the model :param d_k: Dimensionality of queries and keys :param d_v: Dimensionality

:param d_model: Output dimensionality of the model :param d_k: Dimensionality of queries and keys :param d_v: Dimensionality