Spaces:
Running
on
Zero
Running
on
Zero
wondervictor
commited on
Update model.py
Browse files
model.py
CHANGED
@@ -13,6 +13,10 @@ import time
|
|
13 |
from autoregressive.models.generate import generate
|
14 |
from condition.midas.depth import MidasDetector
|
15 |
|
|
|
|
|
|
|
|
|
16 |
models = {
|
17 |
"canny": "checkpoints/canny_MR.safetensors",
|
18 |
"depth": "checkpoints/depth_MR.safetensors",
|
@@ -48,7 +52,8 @@ class Model:
|
|
48 |
self.gpt_model_canny = self.load_gpt(condition_type='canny')
|
49 |
self.gpt_model_depth = self.load_gpt(condition_type='depth')
|
50 |
self.get_control_canny = CannyDetector()
|
51 |
-
self.get_control_depth = MidasDetector('cuda')
|
|
|
52 |
|
53 |
def to(self, device):
|
54 |
self.gpt_model_canny.to('cuda')
|
@@ -196,11 +201,18 @@ class Model:
|
|
196 |
# self.get_control_depth.model.to(self.device)
|
197 |
# self.vq_model.to(self.device)
|
198 |
image_tensor = torch.from_numpy(np.array(image)).to(self.device)
|
199 |
-
condition_img = torch.from_numpy(
|
200 |
-
|
201 |
-
condition_img = condition_img.unsqueeze(0).repeat(2, 3, 1, 1)
|
202 |
-
condition_img = condition_img.to(self.device)
|
203 |
-
condition_img = 2 * (condition_img / 255 - 0.5)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
204 |
prompts = [prompt] * 2
|
205 |
caption_embs, emb_masks = self.t5_model.get_text_embeddings(prompts)
|
206 |
|
|
|
13 |
from autoregressive.models.generate import generate
|
14 |
from condition.midas.depth import MidasDetector
|
15 |
|
16 |
+
from controlnet_aux import (
|
17 |
+
MidasDetector,
|
18 |
+
)
|
19 |
+
|
20 |
models = {
|
21 |
"canny": "checkpoints/canny_MR.safetensors",
|
22 |
"depth": "checkpoints/depth_MR.safetensors",
|
|
|
52 |
self.gpt_model_canny = self.load_gpt(condition_type='canny')
|
53 |
self.gpt_model_depth = self.load_gpt(condition_type='depth')
|
54 |
self.get_control_canny = CannyDetector()
|
55 |
+
# self.get_control_depth = MidasDetector('cuda')
|
56 |
+
self.get_control_depth = MidasDetector.from_pretrained("lllyasviel/Annotators")
|
57 |
|
58 |
def to(self, device):
|
59 |
self.gpt_model_canny.to('cuda')
|
|
|
201 |
# self.get_control_depth.model.to(self.device)
|
202 |
# self.vq_model.to(self.device)
|
203 |
image_tensor = torch.from_numpy(np.array(image)).to(self.device)
|
204 |
+
# condition_img = torch.from_numpy(
|
205 |
+
# self.get_control_depth(image_tensor)).unsqueeze(0)
|
206 |
+
# condition_img = condition_img.unsqueeze(0).repeat(2, 3, 1, 1)
|
207 |
+
# condition_img = condition_img.to(self.device)
|
208 |
+
# condition_img = 2 * (condition_img / 255 - 0.5)
|
209 |
+
|
210 |
+
control_image = self.get_control_depth(
|
211 |
+
image=image,
|
212 |
+
image_resolution=512,
|
213 |
+
detect_resolution=512,
|
214 |
+
)
|
215 |
+
|
216 |
prompts = [prompt] * 2
|
217 |
caption_embs, emb_masks = self.t5_model.get_text_embeddings(prompts)
|
218 |
|