Why is the 2nd argument to f
a tuple rather than just an integer (per the docs here)?
def pool(self, x, bs, is_max):
f = F.adaptive_max_pool1d if is_max else F.adaptive_avg_pool1d
return f(x.permute(1,2,0), (1,)).view(bs,-1)
Why is the 2nd argument to f
a tuple rather than just an integer (per the docs here)?
def pool(self, x, bs, is_max):
f = F.adaptive_max_pool1d if is_max else F.adaptive_avg_pool1d
return f(x.permute(1,2,0), (1,)).view(bs,-1)
Is the adaptive_avg_pool1d function implemented in torch.nn.functional (F) yet? I only see a docstring where the source should be
adaptive_avg_pool1d = _add_docstr(torch.adaptive_avg_pool1d, r"""
adaptive_avg_pool1d(input, output_size) -> TensorApplies a 1D adaptive average pooling over an input signal composed of
several input planes.See :class:
~torch.nn.AdaptiveAvgPool1d
for details and output shape.Args:
output_size: the target output size (single integer)
“”")