(self, features)
| 166 | self.preprocessing_dim = preprocessing_dim |
| 167 | |
| 168 | def forward(self, features): |
| 169 | features = features.reshape(len(features), 1, -1) |
| 170 | return F.adaptive_avg_pool1d(features, self.preprocessing_dim).squeeze(1) |
| 171 | |
| 172 | |
| 173 | class Aggregator(torch.nn.Module): |
nothing calls this directly
no outgoing calls
no test coverage detected