Forward function. Args: x (Tensor): Input tensor with shape (n, c, h, w). Returns: Tensor: Forward results.
(self, x)
| 139 | self.register_buffer('std', torch.Tensor([0.229, 0.224, 0.225]).view(1, 3, 1, 1)) |
| 140 | |
| 141 | def forward(self, x): |
| 142 | """Forward function. |
| 143 | |
| 144 | Args: |
| 145 | x (Tensor): Input tensor with shape (n, c, h, w). |
| 146 | |
| 147 | Returns: |
| 148 | Tensor: Forward results. |
| 149 | """ |
| 150 | if self.range_norm: |
| 151 | x = (x + 1) / 2 |
| 152 | if self.use_input_norm: |
| 153 | x = (x - self.mean) / self.std |
| 154 | output = {} |
| 155 | |
| 156 | for key, layer in self.vgg_net._modules.items(): |
| 157 | x = layer(x) |
| 158 | if key in self.layer_name_list: |
| 159 | output[key] = x.clone() |
| 160 | |
| 161 | return output |
nothing calls this directly
no outgoing calls
no test coverage detected