Base Class

Base Class#

Module#

class Module():

Base class for all neural network modules, inspired by PyTorch’s nn.Module.

Methods#

buffers#

def buffers(self) -> 'Iterator[Tensor]':

compile#

def compile(self, **jit_kwargs):

eval#

def eval(self):

extra_repr#

def extra_repr(self) -> 'str':

forward#

def forward(self, *args, **kwargs):

load_state_dict#

def load_state_dict(self, state_dict: 'OrderedDict[str, Tensor]'):

modules#

def modules(self) -> 'Iterator[Module]':

named_buffers#

def named_buffers(self, prefix: 'str' = '') -> 'Iterator[tuple[str, Tensor]]':

named_parameters#

def named_parameters(self, prefix: 'str' = '') -> 'Iterator[tuple[str, Tensor]]':

parameters#

def parameters(self) -> 'Iterator[Tensor]':

register_buffer#

def register_buffer(self, name: 'str', tensor: 'Tensor | None'):

Adds a persistent buffer to the module.

state_dict#

def state_dict(self) -> 'OrderedDict[str, Tensor]':

train#

def train(self):

zero_grad#

def zero_grad(self) -> 'None':
Contents