Update Flux docstrings #10423
yiyixuxu
merged
1 commit into
main
from
flux-docstrings-update
a-r-r-o-w260 days ago (edited 259 days ago)
- Updates some docs and type hints
- Removes the access of attributes via config in transformer init
- torch.FloatTensor -> torch.Tensor because FloatTensor only exists for backward-compatibility reasons
update
4ccc7329
a-r-r-o-w
commented
on 2025-01-01
src/diffusers/models/transformers/transformer_flux.py
| 259 | 280 | ): | |
| 260 | 281 | super().__init__() | |
| 261 | 282 | self.out_channels = out_channels or in_channels | |
| 262 | self.inner_dim = self.config.num_attention_heads * self.config.attention_head_dim |
a-r-r-o-w260 days agodon't really prefer accessing readily available init attributes via the config. I think we were okay with doing this here
HuggingFaceDocBuilderDev259 days ago
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.
hlky
approved these changes
on 2025-01-02
Login to write a write a comment.