AI/diffusers
1
0
Fork 0
mirror of https://github.com/huggingface/diffusers.git synced 2026-01-29 07:22:12 +03:00
Code Activity

[chore] fix-copies to flux pipelines (#10941)

Browse Source 
fix-copies went uncaught it seems.
This commit is contained in:
Sayak Paul
2025-03-03 11:21:57 +05:30
committed by GitHub
parent 54043c3e2e
commit 4aaa0d21ba
3 changed files with 36 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
src/diffusers/pipelines/flux/pipeline_flux_controlnet.py
View File

@@ -440,23 +440,28 @@ class FluxControlNetPipeline(DiffusionPipeline, FluxLoraLoaderMixin, FromSingleF
if not isinstance(ip_adapter_image, list):
ip_adapter_image = [ip_adapter_image]
if len(ip_adapter_image) != len(self.transformer.encoder_hid_proj.image_projection_layers):
if len(ip_adapter_image) != self.transformer.encoder_hid_proj.num_ip_adapters:
raise ValueError(
f"`ip_adapter_image` must have same length as the number of IP Adapters. Got {len(ip_adapter_image)} images and {len(self.transformer.encoder_hid_proj.image_projection_layers)} IP Adapters."
f"`ip_adapter_image` must have same length as the number of IP Adapters. Got {len(ip_adapter_image)} images and {self.transformer.encoder_hid_proj.num_ip_adapters} IP Adapters."
)
for single_ip_adapter_image, image_proj_layer in zip(
ip_adapter_image, self.transformer.encoder_hid_proj.image_projection_layers
):
for single_ip_adapter_image in ip_adapter_image:
single_image_embeds = self.encode_image(single_ip_adapter_image, device, 1)
image_embeds.append(single_image_embeds[None, :])
else:
if not isinstance(ip_adapter_image_embeds, list):
ip_adapter_image_embeds = [ip_adapter_image_embeds]
if len(ip_adapter_image_embeds) != self.transformer.encoder_hid_proj.num_ip_adapters:
raise ValueError(
f"`ip_adapter_image_embeds` must have same length as the number of IP Adapters. Got {len(ip_adapter_image_embeds)} image embeds and {self.transformer.encoder_hid_proj.num_ip_adapters} IP Adapters."
)
for single_image_embeds in ip_adapter_image_embeds:
image_embeds.append(single_image_embeds)
ip_adapter_image_embeds = []
for i, single_image_embeds in enumerate(image_embeds):
for single_image_embeds in image_embeds:
single_image_embeds = torch.cat([single_image_embeds] * num_images_per_prompt, dim=0)
single_image_embeds = single_image_embeds.to(device=device)
ip_adapter_image_embeds.append(single_image_embeds)

19
src/diffusers/pipelines/flux/pipeline_flux_img2img.py
View File

@@ -427,23 +427,28 @@ class FluxImg2ImgPipeline(DiffusionPipeline, FluxLoraLoaderMixin, FromSingleFile
if not isinstance(ip_adapter_image, list):
ip_adapter_image = [ip_adapter_image]
if len(ip_adapter_image) != len(self.transformer.encoder_hid_proj.image_projection_layers):
if len(ip_adapter_image) != self.transformer.encoder_hid_proj.num_ip_adapters:
raise ValueError(
f"`ip_adapter_image` must have same length as the number of IP Adapters. Got {len(ip_adapter_image)} images and {len(self.transformer.encoder_hid_proj.image_projection_layers)} IP Adapters."
f"`ip_adapter_image` must have same length as the number of IP Adapters. Got {len(ip_adapter_image)} images and {self.transformer.encoder_hid_proj.num_ip_adapters} IP Adapters."
)
for single_ip_adapter_image, image_proj_layer in zip(
ip_adapter_image, self.transformer.encoder_hid_proj.image_projection_layers
):
for single_ip_adapter_image in ip_adapter_image:
single_image_embeds = self.encode_image(single_ip_adapter_image, device, 1)
image_embeds.append(single_image_embeds[None, :])
else:
if not isinstance(ip_adapter_image_embeds, list):
ip_adapter_image_embeds = [ip_adapter_image_embeds]
if len(ip_adapter_image_embeds) != self.transformer.encoder_hid_proj.num_ip_adapters:
raise ValueError(
f"`ip_adapter_image_embeds` must have same length as the number of IP Adapters. Got {len(ip_adapter_image_embeds)} image embeds and {self.transformer.encoder_hid_proj.num_ip_adapters} IP Adapters."
)
for single_image_embeds in ip_adapter_image_embeds:
image_embeds.append(single_image_embeds)
ip_adapter_image_embeds = []
for i, single_image_embeds in enumerate(image_embeds):
for single_image_embeds in image_embeds:
single_image_embeds = torch.cat([single_image_embeds] * num_images_per_prompt, dim=0)
single_image_embeds = single_image_embeds.to(device=device)
ip_adapter_image_embeds.append(single_image_embeds)

19
src/diffusers/pipelines/flux/pipeline_flux_inpaint.py
View File

@@ -432,23 +432,28 @@ class FluxInpaintPipeline(DiffusionPipeline, FluxLoraLoaderMixin, FluxIPAdapterM
if not isinstance(ip_adapter_image, list):
ip_adapter_image = [ip_adapter_image]
if len(ip_adapter_image) != len(self.transformer.encoder_hid_proj.image_projection_layers):
if len(ip_adapter_image) != self.transformer.encoder_hid_proj.num_ip_adapters:
raise ValueError(
f"`ip_adapter_image` must have same length as the number of IP Adapters. Got {len(ip_adapter_image)} images and {len(self.transformer.encoder_hid_proj.image_projection_layers)} IP Adapters."
f"`ip_adapter_image` must have same length as the number of IP Adapters. Got {len(ip_adapter_image)} images and {self.transformer.encoder_hid_proj.num_ip_adapters} IP Adapters."
)
for single_ip_adapter_image, image_proj_layer in zip(
ip_adapter_image, self.transformer.encoder_hid_proj.image_projection_layers
):
for single_ip_adapter_image in ip_adapter_image:
single_image_embeds = self.encode_image(single_ip_adapter_image, device, 1)
image_embeds.append(single_image_embeds[None, :])
else:
if not isinstance(ip_adapter_image_embeds, list):
ip_adapter_image_embeds = [ip_adapter_image_embeds]
if len(ip_adapter_image_embeds) != self.transformer.encoder_hid_proj.num_ip_adapters:
raise ValueError(
f"`ip_adapter_image_embeds` must have same length as the number of IP Adapters. Got {len(ip_adapter_image_embeds)} image embeds and {self.transformer.encoder_hid_proj.num_ip_adapters} IP Adapters."
)
for single_image_embeds in ip_adapter_image_embeds:
image_embeds.append(single_image_embeds)
ip_adapter_image_embeds = []
for i, single_image_embeds in enumerate(image_embeds):
for single_image_embeds in image_embeds:
single_image_embeds = torch.cat([single_image_embeds] * num_images_per_prompt, dim=0)
single_image_embeds = single_image_embeds.to(device=device)
ip_adapter_image_embeds.append(single_image_embeds)