Foreign-Mirrors/Fooocus
1
0
Fork 0
mirror of https://github.com/lllyasviel/Fooocus.git synced 2026-08-16 13:13:16 +02:00
Code Issues Packages Projects Releases Wiki Activity

2.1.844

Browse Source 
* maintain clip vision device
* update links in troubleshoot
This commit is contained in:
lllyasviel
2023-12-16 10:34:47 -08:00
committed by GitHub
parent 776e080b3f
commit 3a727fd240
10 changed files with 113 additions and 77 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ldm_patched/modules/clip_vision.py
+12 -19
View File
@@ -19,11 +19,13 @@ class Output:
def clip_preprocess(image, size=224):
mean = torch.tensor([ 0.48145466,0.4578275,0.40821073], device=image.device, dtype=image.dtype)
std = torch.tensor([0.26862954,0.26130258,0.27577711], device=image.device, dtype=image.dtype)
scale = (size / min(image.shape[1], image.shape[2]))
image = torch.nn.functional.interpolate(image.movedim(-1, 1), size=(round(scale * image.shape[1]), round(scale * image.shape[2])), mode="bicubic", antialias=True)
h = (image.shape[2] - size)//2
w = (image.shape[3] - size)//2
image = image[:,:,h:h+size,w:w+size]
image = image.movedim(-1, 1)
if not (image.shape[2] == size and image.shape[3] == size):
scale = (size / min(image.shape[2], image.shape[3]))
image = torch.nn.functional.interpolate(image, size=(round(scale * image.shape[2]), round(scale * image.shape[3])), mode="bicubic", antialias=True)
h = (image.shape[2] - size)//2
w = (image.shape[3] - size)//2
image = image[:,:,h:h+size,w:w+size]
image = torch.clip((255. * image), 0, 255).round() / 255.0
return (image - mean.view([3,1,1])) / std.view([3,1,1])
@@ -34,11 +36,9 @@ class ClipVisionModel():
self.load_device = ldm_patched.modules.model_management.text_encoder_device()
offload_device = ldm_patched.modules.model_management.text_encoder_offload_device()
self.dtype = torch.float32
if ldm_patched.modules.model_management.should_use_fp16(self.load_device, prioritize_performance=False):
self.dtype = torch.float16
self.model = ldm_patched.modules.clip_model.CLIPVisionModelProjection(config, self.dtype, offload_device, ldm_patched.modules.ops.disable_weight_init)
self.dtype = ldm_patched.modules.model_management.text_encoder_dtype(self.load_device)
self.model = ldm_patched.modules.clip_model.CLIPVisionModelProjection(config, self.dtype, offload_device, ldm_patched.modules.ops.manual_cast)
self.model.eval()
self.patcher = ldm_patched.modules.model_patcher.ModelPatcher(self.model, load_device=self.load_device, offload_device=offload_device)
def load_sd(self, sd):
@@ -46,15 +46,8 @@ class ClipVisionModel():
def encode_image(self, image):
ldm_patched.modules.model_management.load_model_gpu(self.patcher)
pixel_values = clip_preprocess(image.to(self.load_device))
if self.dtype != torch.float32:
precision_scope = torch.autocast
else:
precision_scope = lambda a, b: contextlib.nullcontext(a)
with precision_scope(ldm_patched.modules.model_management.get_autocast_device(self.load_device), torch.float32):
out = self.model(pixel_values=pixel_values, intermediate_output=-2)
pixel_values = clip_preprocess(image.to(self.load_device)).float()
out = self.model(pixel_values=pixel_values, intermediate_output=-2)
outputs = Output()
outputs["last_hidden_state"] = out[0].to(ldm_patched.modules.model_management.intermediate_device())