| 222 | |
| 223 | |
| 224 | def forward(self, sample): |
| 225 | sample = sample / self.scaling_factor |
| 226 | hidden_states = self.conv_in(sample) |
| 227 | |
| 228 | for block in self.blocks: |
| 229 | hidden_states = block(hidden_states, sample) |
| 230 | |
| 231 | hidden_states = self.norm_out(hidden_states, sample) |
| 232 | hidden_states = self.conv_act(hidden_states) |
| 233 | hidden_states = self.conv_out(hidden_states) |
| 234 | |
| 235 | return hidden_states |
| 236 | |
| 237 | |
| 238 | def decode_video(self, sample, tiled=True, tile_size=(60, 90), tile_stride=(30, 45), progress_bar=lambda x:x): |