| 23 | """MLP as used in Vision Transformer, MLP-Mixer and related networks""" |
| 24 | |
| 25 | def __init__( |
| 26 | self, |
| 27 | in_channels, |
| 28 | hidden_channels=None, |
| 29 | out_features=None, |
| 30 | act_layer=nn.GELU, |
| 31 | norm_layer=None, |
| 32 | bias=True, |
| 33 | drop=0.0, |
| 34 | use_conv=False, |
| 35 | device=None, |
| 36 | dtype=None, |
| 37 | ): |
| 38 | super().__init__() |
| 39 | out_features = out_features or in_channels |
| 40 | hidden_channels = hidden_channels or in_channels |
| 41 | bias = (bias, bias) |
| 42 | drop_probs = (drop, drop) |
| 43 | linear_layer = partial(nn.Conv2d, kernel_size=1) if use_conv else nn.Linear |
| 44 | |
| 45 | self.fc1 = linear_layer( |
| 46 | in_channels, hidden_channels, bias=bias[0], device=device, dtype=dtype |
| 47 | ) |
| 48 | self.act = act_layer() |
| 49 | self.drop1 = nn.Dropout(drop_probs[0]) |
| 50 | self.norm = ( |
| 51 | norm_layer(hidden_channels, device=device, dtype=dtype) |
| 52 | if norm_layer is not None |
| 53 | else nn.Identity() |
| 54 | ) |
| 55 | self.fc2 = linear_layer( |
| 56 | hidden_channels, out_features, bias=bias[1], device=device, dtype=dtype |
| 57 | ) |
| 58 | self.drop2 = nn.Dropout(drop_probs[1]) |
| 59 | |
| 60 | def forward(self, x): |
| 61 | x = self.fc1(x) |