Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -2,7 +2,6 @@ import sys
|
|
2 |
import os
|
3 |
from pathlib import Path
|
4 |
import gc
|
5 |
-
import traceback
|
6 |
|
7 |
# Add the StableCascade and CSD directories to the Python path
|
8 |
app_dir = Path(__file__).parent
|
@@ -28,7 +27,6 @@ from utils import WurstCoreCRBM
|
|
28 |
from gdf.schedulers import CosineSchedule
|
29 |
from gdf import VPScaler, CosineTNoiseCond, DDPMSampler, P2LossWeight, AdaptiveLossWeight
|
30 |
from gdf.targets import EpsilonTarget
|
31 |
-
import PIL
|
32 |
|
33 |
# Enable mixed precision
|
34 |
torch.backends.cuda.matmul.allow_tf32 = True
|
@@ -75,160 +73,171 @@ if low_vram:
|
|
75 |
|
76 |
clear_gpu_cache()
|
77 |
|
78 |
-
#
|
79 |
config_file = 'third_party/StableCascade/configs/inference/stage_c_3b.yaml'
|
80 |
with open(config_file, "r", encoding="utf-8") as file:
|
81 |
loaded_config = yaml.safe_load(file)
|
82 |
|
|
|
|
|
|
|
83 |
config_file_b = 'third_party/StableCascade/configs/inference/stage_b_3b.yaml'
|
84 |
with open(config_file_b, "r", encoding="utf-8") as file:
|
85 |
config_file_b = yaml.safe_load(file)
|
86 |
-
|
87 |
-
def initialize_models():
|
88 |
-
global models_rbm, models_b, extras, extras_b, core, core_b
|
89 |
-
|
90 |
-
# Clear any existing models from memory
|
91 |
-
models_rbm = None
|
92 |
-
models_b = None
|
93 |
-
extras = None
|
94 |
-
extras_b = None
|
95 |
-
|
96 |
-
# Clear GPU cache
|
97 |
-
clear_gpu_cache()
|
98 |
-
|
99 |
-
# Initialize models
|
100 |
-
core = WurstCoreCRBM(config_dict=loaded_config, device=device, training=False)
|
101 |
-
core_b = WurstCoreB(config_dict=config_file_b, device=device, training=False)
|
102 |
-
|
103 |
-
extras = core.setup_extras_pre()
|
104 |
-
models = core.setup_models(extras)
|
105 |
-
|
106 |
-
extras_b = core_b.setup_extras_pre()
|
107 |
-
models_b = core_b.setup_models(extras_b, skip_clip=True)
|
108 |
-
models_b = WurstCoreB.Models(
|
109 |
-
**{**models_b.to_dict(), 'tokenizer': models.tokenizer, 'text_model': models.text_model}
|
110 |
-
)
|
111 |
-
|
112 |
-
# Initialize models_rbm
|
113 |
-
generator_rbm = StageCRBM()
|
114 |
-
for param_name, param in load_or_fail(core.config.generator_checkpoint_path).items():
|
115 |
-
set_module_tensor_to_device(generator_rbm, param_name, "cpu", value=param)
|
116 |
-
|
117 |
-
generator_rbm = generator_rbm.to(getattr(torch, core.config.dtype)).to(device)
|
118 |
-
generator_rbm = core.load_model(generator_rbm, 'generator')
|
119 |
-
|
120 |
-
models_rbm = core.Models(
|
121 |
-
effnet=models.effnet,
|
122 |
-
previewer=models.previewer,
|
123 |
-
generator=generator_rbm,
|
124 |
-
generator_ema=models.generator_ema,
|
125 |
-
tokenizer=models.tokenizer,
|
126 |
-
text_model=models.text_model,
|
127 |
-
image_model=models.image_model
|
128 |
-
)
|
129 |
-
|
130 |
-
# Move models to appropriate devices
|
131 |
-
models_rbm.generator.to(device).eval().requires_grad_(False)
|
132 |
-
models_b.generator.to(device).eval().requires_grad_(False)
|
133 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
134 |
clear_gpu_cache()
|
135 |
|
136 |
-
|
137 |
-
|
138 |
-
|
139 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
140 |
|
141 |
-
|
142 |
-
|
143 |
-
models_b.to(device)
|
144 |
|
145 |
-
|
146 |
-
|
147 |
-
|
148 |
-
|
149 |
-
|
150 |
-
|
151 |
-
|
152 |
-
|
153 |
-
|
154 |
-
|
155 |
-
|
156 |
-
|
157 |
-
|
158 |
-
|
159 |
-
|
160 |
-
|
161 |
-
|
162 |
-
|
163 |
-
|
164 |
-
|
165 |
-
|
166 |
-
|
167 |
-
|
168 |
-
|
169 |
-
|
170 |
-
|
171 |
-
|
172 |
-
|
173 |
-
|
174 |
-
|
175 |
-
|
176 |
-
|
177 |
-
|
178 |
-
|
179 |
-
|
180 |
-
|
181 |
-
|
182 |
-
|
183 |
-
|
184 |
-
|
185 |
-
|
186 |
-
|
187 |
-
|
188 |
-
|
189 |
-
|
190 |
-
|
191 |
-
|
192 |
-
|
193 |
-
|
194 |
-
|
195 |
-
|
196 |
-
|
197 |
-
|
198 |
-
|
199 |
-
|
200 |
-
sampled = models_b.stage_a.decode(sampled_b).float()
|
201 |
-
|
202 |
-
# Post-process and save the image
|
203 |
-
sampled = sampled.cpu() # Move to CPU before processing
|
204 |
-
|
205 |
-
# Ensure the tensor is in [C, H, W] format
|
206 |
-
if sampled.dim() == 4 and sampled.size(0) == 1:
|
207 |
-
sampled = sampled.squeeze(0)
|
208 |
|
209 |
-
|
210 |
-
|
211 |
-
|
212 |
-
|
213 |
-
|
214 |
-
|
215 |
-
|
216 |
-
|
217 |
-
|
218 |
-
|
219 |
-
|
220 |
-
|
221 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
222 |
|
223 |
return output_file # Return the path to the saved image
|
224 |
|
225 |
import gradio as gr
|
226 |
|
227 |
-
def gradio_interface(style_description, ref_style_file, caption):
|
228 |
-
return infer(style_description, ref_style_file, caption)
|
229 |
-
|
230 |
gr.Interface(
|
231 |
-
fn=
|
232 |
inputs=[gr.Textbox(label="style description"), gr.Image(label="Ref Style File", type="filepath"), gr.Textbox(label="caption")],
|
233 |
outputs=[gr.Image()]
|
234 |
).launch()
|
|
|
2 |
import os
|
3 |
from pathlib import Path
|
4 |
import gc
|
|
|
5 |
|
6 |
# Add the StableCascade and CSD directories to the Python path
|
7 |
app_dir = Path(__file__).parent
|
|
|
27 |
from gdf.schedulers import CosineSchedule
|
28 |
from gdf import VPScaler, CosineTNoiseCond, DDPMSampler, P2LossWeight, AdaptiveLossWeight
|
29 |
from gdf.targets import EpsilonTarget
|
|
|
30 |
|
31 |
# Enable mixed precision
|
32 |
torch.backends.cuda.matmul.allow_tf32 = True
|
|
|
73 |
|
74 |
clear_gpu_cache()
|
75 |
|
76 |
+
# Stage C model configuration
|
77 |
config_file = 'third_party/StableCascade/configs/inference/stage_c_3b.yaml'
|
78 |
with open(config_file, "r", encoding="utf-8") as file:
|
79 |
loaded_config = yaml.safe_load(file)
|
80 |
|
81 |
+
core = WurstCoreCRBM(config_dict=loaded_config, device=device, training=False)
|
82 |
+
|
83 |
+
# Stage B model configuration
|
84 |
config_file_b = 'third_party/StableCascade/configs/inference/stage_b_3b.yaml'
|
85 |
with open(config_file_b, "r", encoding="utf-8") as file:
|
86 |
config_file_b = yaml.safe_load(file)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
87 |
|
88 |
+
core_b = WurstCoreB(config_dict=config_file_b, device=device, training=False)
|
89 |
+
|
90 |
+
# Setup extras and models for Stage C
|
91 |
+
extras = core.setup_extras_pre()
|
92 |
+
|
93 |
+
gdf_rbm = RBM(
|
94 |
+
schedule=CosineSchedule(clamp_range=[0.0001, 0.9999]),
|
95 |
+
input_scaler=VPScaler(), target=EpsilonTarget(),
|
96 |
+
noise_cond=CosineTNoiseCond(),
|
97 |
+
loss_weight=AdaptiveLossWeight(),
|
98 |
+
)
|
99 |
+
|
100 |
+
sampling_configs = {
|
101 |
+
"cfg": 5,
|
102 |
+
"sampler": DDPMSampler(gdf_rbm),
|
103 |
+
"shift": 1,
|
104 |
+
"timesteps": 20
|
105 |
+
}
|
106 |
+
|
107 |
+
extras = core.Extras(
|
108 |
+
gdf=gdf_rbm,
|
109 |
+
sampling_configs=sampling_configs,
|
110 |
+
transforms=extras.transforms,
|
111 |
+
effnet_preprocess=extras.effnet_preprocess,
|
112 |
+
clip_preprocess=extras.clip_preprocess
|
113 |
+
)
|
114 |
+
|
115 |
+
models = core.setup_models(extras)
|
116 |
+
models.generator.eval().requires_grad_(False)
|
117 |
+
|
118 |
+
# Setup extras and models for Stage B
|
119 |
+
extras_b = core_b.setup_extras_pre()
|
120 |
+
models_b = core_b.setup_models(extras_b, skip_clip=True)
|
121 |
+
models_b = WurstCoreB.Models(
|
122 |
+
**{**models_b.to_dict(), 'tokenizer': models.tokenizer, 'text_model': models.text_model}
|
123 |
+
)
|
124 |
+
models_b.generator.bfloat16().eval().requires_grad_(False)
|
125 |
+
|
126 |
+
# Off-load old generator (low VRAM mode)
|
127 |
+
if low_vram:
|
128 |
+
models.generator.to("cpu")
|
129 |
clear_gpu_cache()
|
130 |
|
131 |
+
# Load and configure new generator
|
132 |
+
generator_rbm = StageCRBM()
|
133 |
+
for param_name, param in load_or_fail(core.config.generator_checkpoint_path).items():
|
134 |
+
set_module_tensor_to_device(generator_rbm, param_name, "cpu", value=param)
|
135 |
+
|
136 |
+
generator_rbm = generator_rbm.to(getattr(torch, core.config.dtype)).to(device)
|
137 |
+
generator_rbm = core.load_model(generator_rbm, 'generator')
|
138 |
+
|
139 |
+
# Create models_rbm instance
|
140 |
+
models_rbm = core.Models(
|
141 |
+
effnet=models.effnet,
|
142 |
+
previewer=models.previewer,
|
143 |
+
generator=generator_rbm,
|
144 |
+
generator_ema=models.generator_ema,
|
145 |
+
tokenizer=models.tokenizer,
|
146 |
+
text_model=models.text_model,
|
147 |
+
image_model=models.image_model
|
148 |
+
)
|
149 |
+
models_rbm.generator.eval().requires_grad_(False)
|
150 |
|
151 |
+
def infer(style_description, ref_style_file, caption):
|
152 |
+
clear_gpu_cache() # Clear cache before inference
|
|
|
153 |
|
154 |
+
height=1024
|
155 |
+
width=1024
|
156 |
+
batch_size=1
|
157 |
+
output_file='output.png'
|
158 |
+
|
159 |
+
stage_c_latent_shape, stage_b_latent_shape = calculate_latent_sizes(height, width, batch_size=batch_size)
|
160 |
+
|
161 |
+
extras.sampling_configs['cfg'] = 4
|
162 |
+
extras.sampling_configs['shift'] = 2
|
163 |
+
extras.sampling_configs['timesteps'] = 20
|
164 |
+
extras.sampling_configs['t_start'] = 1.0
|
165 |
+
|
166 |
+
extras_b.sampling_configs['cfg'] = 1.1
|
167 |
+
extras_b.sampling_configs['shift'] = 1
|
168 |
+
extras_b.sampling_configs['timesteps'] = 10
|
169 |
+
extras_b.sampling_configs['t_start'] = 1.0
|
170 |
+
|
171 |
+
ref_style = resize_image(PIL.Image.open(ref_style_file).convert("RGB")).unsqueeze(0).expand(batch_size, -1, -1, -1).to(device)
|
172 |
+
|
173 |
+
batch = {'captions': [caption] * batch_size}
|
174 |
+
batch['style'] = ref_style
|
175 |
+
|
176 |
+
x0_style_forward = models_rbm.effnet(extras.effnet_preprocess(ref_style.to(device)))
|
177 |
+
|
178 |
+
conditions = core.get_conditions(batch, models_rbm, extras, is_eval=True, is_unconditional=False, eval_image_embeds=True, eval_style=True, eval_csd=False)
|
179 |
+
unconditions = core.get_conditions(batch, models_rbm, extras, is_eval=True, is_unconditional=True, eval_image_embeds=False)
|
180 |
+
conditions_b = core_b.get_conditions(batch, models_b, extras_b, is_eval=True, is_unconditional=False)
|
181 |
+
unconditions_b = core_b.get_conditions(batch, models_b, extras_b, is_eval=True, is_unconditional=True)
|
182 |
+
|
183 |
+
if low_vram:
|
184 |
+
# The sampling process uses more vram, so we offload everything except two modules to the cpu.
|
185 |
+
models_to(models_rbm, device="cpu", excepts=["generator", "previewer"])
|
186 |
+
|
187 |
+
# Stage C reverse process.
|
188 |
+
with torch.cuda.amp.autocast(): # Use mixed precision
|
189 |
+
sampling_c = extras.gdf.sample(
|
190 |
+
models_rbm.generator, conditions, stage_c_latent_shape,
|
191 |
+
unconditions, device=device,
|
192 |
+
**extras.sampling_configs,
|
193 |
+
x0_style_forward=x0_style_forward,
|
194 |
+
apply_pushforward=False, tau_pushforward=8,
|
195 |
+
num_iter=3, eta=0.1, tau=20, eval_csd=True,
|
196 |
+
extras=extras, models=models_rbm,
|
197 |
+
lam_style=1, lam_txt_alignment=1.0,
|
198 |
+
use_ddim_sampler=True,
|
199 |
+
)
|
200 |
+
for (sampled_c, _, _) in tqdm(sampling_c, total=extras.sampling_configs['timesteps']):
|
201 |
+
sampled_c = sampled_c
|
202 |
+
|
203 |
+
clear_gpu_cache() # Clear cache between stages
|
204 |
+
|
205 |
+
# Stage B reverse process.
|
206 |
+
with torch.no_grad(), torch.cuda.amp.autocast(dtype=torch.bfloat16):
|
207 |
+
conditions_b['effnet'] = sampled_c
|
208 |
+
unconditions_b['effnet'] = torch.zeros_like(sampled_c)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
209 |
|
210 |
+
sampling_b = extras_b.gdf.sample(
|
211 |
+
models_b.generator, conditions_b, stage_b_latent_shape,
|
212 |
+
unconditions_b, device=device, **extras_b.sampling_configs,
|
213 |
+
)
|
214 |
+
for (sampled_b, _, _) in tqdm(sampling_b, total=extras_b.sampling_configs['timesteps']):
|
215 |
+
sampled_b = sampled_b
|
216 |
+
sampled = models_b.stage_a.decode(sampled_b).float()
|
217 |
+
|
218 |
+
sampled = torch.cat([
|
219 |
+
torch.nn.functional.interpolate(ref_style.cpu(), size=(height, width)),
|
220 |
+
sampled.cpu(),
|
221 |
+
], dim=0)
|
222 |
+
|
223 |
+
# Remove the batch dimension and keep only the generated image
|
224 |
+
sampled = sampled[1] # This selects the generated image, discarding the reference style image
|
225 |
+
|
226 |
+
# Ensure the tensor is in [C, H, W] format
|
227 |
+
if sampled.dim() == 3 and sampled.shape[0] == 3:
|
228 |
+
sampled_image = T.ToPILImage()(sampled) # Convert tensor to PIL image
|
229 |
+
sampled_image.save(output_file) # Save the image as a PNG
|
230 |
+
else:
|
231 |
+
raise ValueError(f"Expected tensor of shape [3, H, W] but got {sampled.shape}")
|
232 |
+
|
233 |
+
clear_gpu_cache() # Clear cache after inference
|
234 |
|
235 |
return output_file # Return the path to the saved image
|
236 |
|
237 |
import gradio as gr
|
238 |
|
|
|
|
|
|
|
239 |
gr.Interface(
|
240 |
+
fn = infer,
|
241 |
inputs=[gr.Textbox(label="style description"), gr.Image(label="Ref Style File", type="filepath"), gr.Textbox(label="caption")],
|
242 |
outputs=[gr.Image()]
|
243 |
).launch()
|