Spaces:
Running
on
Zero
Running
on
Zero
RageshAntony
commited on
added deep cleanup
Browse files
app.py
CHANGED
@@ -8,6 +8,13 @@ from diffusers import (
|
|
8 |
LuminaText2ImgPipeline
|
9 |
)
|
10 |
import spaces
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
|
12 |
# Constants
|
13 |
MAX_SEED = np.iinfo(np.int32).max
|
@@ -47,18 +54,85 @@ MODEL_CONFIGS = {
|
|
47 |
}
|
48 |
}
|
49 |
|
50 |
-
#
|
51 |
pipes = {}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
52 |
|
53 |
def load_pipeline(model_name):
|
|
|
|
|
54 |
config = MODEL_CONFIGS[model_name]
|
|
|
55 |
pipe = config["pipeline_class"].from_pretrained(
|
56 |
config["repo_id"],
|
57 |
torch_dtype=TORCH_DTYPE
|
58 |
)
|
59 |
pipe = pipe.to(DEVICE)
|
|
|
60 |
if hasattr(pipe, 'enable_model_cpu_offload'):
|
61 |
pipe.enable_model_cpu_offload()
|
|
|
|
|
|
|
|
|
62 |
return pipe
|
63 |
|
64 |
@spaces.GPU(duration=180)
|
@@ -74,33 +148,48 @@ def generate_image(
|
|
74 |
num_inference_steps=40,
|
75 |
progress=gr.Progress(track_tqdm=True)
|
76 |
):
|
77 |
-
|
78 |
-
|
79 |
-
|
80 |
-
|
81 |
-
|
82 |
-
|
83 |
-
|
84 |
-
|
85 |
-
|
86 |
-
|
87 |
-
|
88 |
-
|
89 |
-
|
90 |
-
|
91 |
-
|
92 |
-
|
93 |
-
|
94 |
-
|
95 |
-
|
96 |
-
|
97 |
-
|
98 |
-
|
99 |
-
|
100 |
-
|
101 |
-
|
102 |
-
|
103 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
104 |
|
105 |
# Gradio Interface
|
106 |
css = """
|
@@ -173,6 +262,9 @@ with gr.Blocks(css=css) as demo:
|
|
173 |
value=40,
|
174 |
)
|
175 |
|
|
|
|
|
|
|
176 |
# Create tabs for each model
|
177 |
with gr.Tabs() as tabs:
|
178 |
results = {}
|
@@ -188,6 +280,14 @@ with gr.Blocks(css=css) as demo:
|
|
188 |
]
|
189 |
gr.Examples(examples=examples, inputs=[prompt])
|
190 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
191 |
# Handle generation for each model
|
192 |
def generate_all(prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps, progress=gr.Progress()):
|
193 |
outputs = []
|
@@ -199,9 +299,14 @@ with gr.Blocks(css=css) as demo:
|
|
199 |
num_inference_steps, progress
|
200 |
)
|
201 |
outputs.extend([image, used_seed])
|
|
|
|
|
|
|
|
|
202 |
except Exception as e:
|
203 |
outputs.extend([None, None])
|
204 |
print(f"Error generating with {model_name}: {str(e)}")
|
|
|
205 |
return outputs
|
206 |
|
207 |
# Set up the generation trigger
|
|
|
8 |
LuminaText2ImgPipeline
|
9 |
)
|
10 |
import spaces
|
11 |
+
import gc
|
12 |
+
import os
|
13 |
+
import psutil
|
14 |
+
import threading
|
15 |
+
from pathlib import Path
|
16 |
+
import shutil
|
17 |
+
import time
|
18 |
|
19 |
# Constants
|
20 |
MAX_SEED = np.iinfo(np.int32).max
|
|
|
54 |
}
|
55 |
}
|
56 |
|
57 |
+
# Dictionary to store model pipelines
|
58 |
pipes = {}
|
59 |
+
model_locks = {model_name: threading.Lock() for model_name in MODEL_CONFIGS.keys()}
|
60 |
+
|
61 |
+
def get_process_memory():
|
62 |
+
"""Get memory usage of current process in GB"""
|
63 |
+
process = psutil.Process(os.getpid())
|
64 |
+
return process.memory_info().rss / 1024 / 1024 / 1024
|
65 |
+
|
66 |
+
def clear_torch_cache():
|
67 |
+
"""Clear PyTorch's CUDA cache"""
|
68 |
+
if torch.cuda.is_available():
|
69 |
+
torch.cuda.empty_cache()
|
70 |
+
torch.cuda.ipc_collect()
|
71 |
+
|
72 |
+
def remove_cache_dir(model_name):
|
73 |
+
"""Remove the model's cache directory"""
|
74 |
+
cache_dir = Path.home() / '.cache' / 'huggingface' / 'diffusers' / MODEL_CONFIGS[model_name]['repo_id'].replace('/', '--')
|
75 |
+
if cache_dir.exists():
|
76 |
+
shutil.rmtree(cache_dir, ignore_errors=True)
|
77 |
+
|
78 |
+
def deep_cleanup(model_name, pipe):
|
79 |
+
"""Perform deep cleanup of model resources"""
|
80 |
+
try:
|
81 |
+
# 1. Move model to CPU first (helps prevent CUDA memory fragmentation)
|
82 |
+
if hasattr(pipe, 'to'):
|
83 |
+
pipe.to('cpu')
|
84 |
+
|
85 |
+
# 2. Delete all model components explicitly
|
86 |
+
for attr_name in list(pipe.__dict__.keys()):
|
87 |
+
if hasattr(pipe, attr_name):
|
88 |
+
delattr(pipe, attr_name)
|
89 |
+
|
90 |
+
# 3. Remove from pipes dictionary
|
91 |
+
if model_name in pipes:
|
92 |
+
del pipes[model_name]
|
93 |
+
|
94 |
+
# 4. Clear CUDA cache
|
95 |
+
clear_torch_cache()
|
96 |
+
|
97 |
+
# 5. Run garbage collection multiple times
|
98 |
+
for _ in range(3):
|
99 |
+
gc.collect()
|
100 |
+
|
101 |
+
# 6. Remove cached files
|
102 |
+
remove_cache_dir(model_name)
|
103 |
+
|
104 |
+
# 7. Additional CUDA cleanup if available
|
105 |
+
if torch.cuda.is_available():
|
106 |
+
torch.cuda.synchronize()
|
107 |
+
|
108 |
+
# 8. Wait a small amount of time to ensure cleanup
|
109 |
+
time.sleep(1)
|
110 |
+
|
111 |
+
except Exception as e:
|
112 |
+
print(f"Error during cleanup of {model_name}: {str(e)}")
|
113 |
+
|
114 |
+
finally:
|
115 |
+
# Final garbage collection
|
116 |
+
gc.collect()
|
117 |
+
clear_torch_cache()
|
118 |
|
119 |
def load_pipeline(model_name):
|
120 |
+
"""Load model pipeline with memory tracking"""
|
121 |
+
initial_memory = get_process_memory()
|
122 |
config = MODEL_CONFIGS[model_name]
|
123 |
+
|
124 |
pipe = config["pipeline_class"].from_pretrained(
|
125 |
config["repo_id"],
|
126 |
torch_dtype=TORCH_DTYPE
|
127 |
)
|
128 |
pipe = pipe.to(DEVICE)
|
129 |
+
|
130 |
if hasattr(pipe, 'enable_model_cpu_offload'):
|
131 |
pipe.enable_model_cpu_offload()
|
132 |
+
|
133 |
+
final_memory = get_process_memory()
|
134 |
+
print(f"Memory used by {model_name}: {final_memory - initial_memory:.2f} GB")
|
135 |
+
|
136 |
return pipe
|
137 |
|
138 |
@spaces.GPU(duration=180)
|
|
|
148 |
num_inference_steps=40,
|
149 |
progress=gr.Progress(track_tqdm=True)
|
150 |
):
|
151 |
+
with model_locks[model_name]:
|
152 |
+
try:
|
153 |
+
progress(0, desc=f"Loading {model_name} model...")
|
154 |
+
|
155 |
+
# Load model if not already loaded
|
156 |
+
if model_name not in pipes:
|
157 |
+
pipes[model_name] = load_pipeline(model_name)
|
158 |
+
|
159 |
+
pipe = pipes[model_name]
|
160 |
+
|
161 |
+
if randomize_seed:
|
162 |
+
seed = random.randint(0, MAX_SEED)
|
163 |
+
|
164 |
+
generator = torch.Generator(DEVICE).manual_seed(seed)
|
165 |
+
|
166 |
+
progress(0.3, desc=f"Generating image with {model_name}...")
|
167 |
+
|
168 |
+
# Generate image
|
169 |
+
image = pipe(
|
170 |
+
prompt=prompt,
|
171 |
+
negative_prompt=negative_prompt,
|
172 |
+
guidance_scale=guidance_scale,
|
173 |
+
num_inference_steps=num_inference_steps,
|
174 |
+
width=width,
|
175 |
+
height=height,
|
176 |
+
generator=generator,
|
177 |
+
).images[0]
|
178 |
+
|
179 |
+
progress(0.9, desc=f"Cleaning up {model_name} resources...")
|
180 |
+
|
181 |
+
# Cleanup after generation
|
182 |
+
deep_cleanup(model_name, pipe)
|
183 |
+
|
184 |
+
progress(1.0, desc=f"Generation complete with {model_name}")
|
185 |
+
return image, seed
|
186 |
+
|
187 |
+
except Exception as e:
|
188 |
+
print(f"Error with {model_name}: {str(e)}")
|
189 |
+
# Ensure cleanup happens even if generation fails
|
190 |
+
if model_name in pipes:
|
191 |
+
deep_cleanup(model_name, pipes[model_name])
|
192 |
+
raise e
|
193 |
|
194 |
# Gradio Interface
|
195 |
css = """
|
|
|
262 |
value=40,
|
263 |
)
|
264 |
|
265 |
+
# Memory usage indicator
|
266 |
+
memory_indicator = gr.Markdown("Current memory usage: 0 GB")
|
267 |
+
|
268 |
# Create tabs for each model
|
269 |
with gr.Tabs() as tabs:
|
270 |
results = {}
|
|
|
280 |
]
|
281 |
gr.Examples(examples=examples, inputs=[prompt])
|
282 |
|
283 |
+
def update_memory_usage():
|
284 |
+
"""Update memory usage display"""
|
285 |
+
memory_gb = get_process_memory()
|
286 |
+
if torch.cuda.is_available():
|
287 |
+
cuda_memory_gb = torch.cuda.memory_allocated() / 1024 / 1024 / 1024
|
288 |
+
return f"Current memory usage: System RAM: {memory_gb:.2f} GB, CUDA: {cuda_memory_gb:.2f} GB"
|
289 |
+
return f"Current memory usage: System RAM: {memory_gb:.2f} GB"
|
290 |
+
|
291 |
# Handle generation for each model
|
292 |
def generate_all(prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps, progress=gr.Progress()):
|
293 |
outputs = []
|
|
|
299 |
num_inference_steps, progress
|
300 |
)
|
301 |
outputs.extend([image, used_seed])
|
302 |
+
|
303 |
+
# Update memory usage after each model
|
304 |
+
memory_indicator.update(update_memory_usage())
|
305 |
+
|
306 |
except Exception as e:
|
307 |
outputs.extend([None, None])
|
308 |
print(f"Error generating with {model_name}: {str(e)}")
|
309 |
+
|
310 |
return outputs
|
311 |
|
312 |
# Set up the generation trigger
|