Eli Simhayev

Results 31 comments of Eli Simhayev

@ezyang, to clarify, are you saying that for x with shape NCHW ```python F.layer_norm(x.permute(0, 2, 3, 1), self.normalized_shape, self.weight, self.bias, self.eps).permute(0, 3, 1, 2) ``` might perform fast in PT2...