MCPcopy Create free account
hub / github.com/tdrussell/diffusion-pipe / forward

Method forward

models/auraflow.py:227–236  ·  view source on GitHub ↗
(self, inputs)

Source from the content-addressed store, hash-verified

225
226 @torch.autocast('cuda', dtype=AUTOCAST_DTYPE)
227 def forward(self, inputs):
228 hidden_states, encoder_hidden_states, temb, height_width = inputs
229
230 encoder_hidden_states, hidden_states = self.block(
231 hidden_states=hidden_states,
232 encoder_hidden_states=encoder_hidden_states,
233 temb=temb,
234 )
235
236 return make_contiguous(hidden_states, encoder_hidden_states, temb, height_width)
237
238
239def concat_hidden_states(inputs):

Callers

nothing calls this directly

Calls 1

make_contiguousFunction · 0.90

Tested by

no test coverage detected