Spaces:
Runtime error
Runtime error
feifeiobama
commited on
Commit
•
b664a31
1
Parent(s):
d36add3
Update app.py
Browse filesChange the demo to 384p
app.py
CHANGED
@@ -10,15 +10,12 @@ from diffusers.utils import export_to_video
|
|
10 |
import spaces
|
11 |
import uuid
|
12 |
|
13 |
-
import subprocess
|
14 |
-
subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
|
15 |
-
|
16 |
is_canonical = True if os.environ.get("SPACE_ID") == "Pyramid-Flow/pyramid-flow" else False
|
17 |
|
18 |
# Constants
|
19 |
MODEL_PATH = "pyramid-flow-model"
|
20 |
MODEL_REPO = "rain1011/pyramid-flow-sd3"
|
21 |
-
MODEL_VARIANT = "
|
22 |
MODEL_DTYPE = "bf16"
|
23 |
|
24 |
def center_crop(image, target_width, target_height):
|
@@ -66,19 +63,18 @@ model = load_model()
|
|
66 |
# Text-to-video generation function
|
67 |
@spaces.GPU(duration=120)
|
68 |
def generate_video(prompt, image=None, duration=5, guidance_scale=9, video_guidance_scale=5, progress=gr.Progress(track_tqdm=True)):
|
69 |
-
multiplier =
|
70 |
-
temp = int(duration *
|
71 |
torch_dtype = torch.bfloat16 if MODEL_DTYPE == "bf16" else torch.float32
|
72 |
if(image):
|
73 |
-
cropped_image = center_crop(image,
|
74 |
-
resized_image = cropped_image.resize((
|
75 |
with torch.no_grad(), torch.cuda.amp.autocast(enabled=True, dtype=torch_dtype):
|
76 |
frames = model.generate_i2v(
|
77 |
prompt=prompt,
|
78 |
input_image=resized_image,
|
79 |
num_inference_steps=[10, 10, 10],
|
80 |
temp=temp,
|
81 |
-
guidance_scale=7.0,
|
82 |
video_guidance_scale=video_guidance_scale,
|
83 |
output_type="pil",
|
84 |
save_memory=True,
|
@@ -89,8 +85,8 @@ def generate_video(prompt, image=None, duration=5, guidance_scale=9, video_guida
|
|
89 |
prompt=prompt,
|
90 |
num_inference_steps=[20, 20, 20],
|
91 |
video_num_inference_steps=[10, 10, 10],
|
92 |
-
height=
|
93 |
-
width=
|
94 |
temp=temp,
|
95 |
guidance_scale=guidance_scale,
|
96 |
video_guidance_scale=video_guidance_scale,
|
@@ -98,14 +94,14 @@ def generate_video(prompt, image=None, duration=5, guidance_scale=9, video_guida
|
|
98 |
save_memory=True,
|
99 |
)
|
100 |
output_path = f"{str(uuid.uuid4())}_output_video.mp4"
|
101 |
-
export_to_video(frames, output_path, fps=
|
102 |
return output_path
|
103 |
|
104 |
# Gradio interface
|
105 |
with gr.Blocks() as demo:
|
106 |
-
gr.Markdown("# Pyramid Flow")
|
107 |
-
gr.Markdown("Pyramid Flow is a training-efficient Autoregressive Video Generation model based on Flow Matching
|
108 |
-
gr.Markdown("[[Paper](https://arxiv.org/pdf/2410.05954)], [[Model](https://huggingface.co/rain1011/pyramid-flow-sd3)], [[Code](https://github.com/jy0205/Pyramid-Flow)]")
|
109 |
|
110 |
with gr.Row():
|
111 |
with gr.Column():
|
@@ -113,8 +109,8 @@ with gr.Blocks() as demo:
|
|
113 |
i2v_image = gr.Image(type="pil", label="Input Image")
|
114 |
t2v_prompt = gr.Textbox(label="Prompt")
|
115 |
with gr.Accordion("Advanced settings", open=False):
|
116 |
-
t2v_duration = gr.Slider(minimum=1, maximum=
|
117 |
-
t2v_guidance_scale = gr.Slider(minimum=1, maximum=15, value=
|
118 |
t2v_video_guidance_scale = gr.Slider(minimum=1, maximum=15, value=5, step=0.1, label="Video Guidance Scale")
|
119 |
t2v_generate_btn = gr.Button("Generate Video")
|
120 |
with gr.Column():
|
|
|
10 |
import spaces
|
11 |
import uuid
|
12 |
|
|
|
|
|
|
|
13 |
is_canonical = True if os.environ.get("SPACE_ID") == "Pyramid-Flow/pyramid-flow" else False
|
14 |
|
15 |
# Constants
|
16 |
MODEL_PATH = "pyramid-flow-model"
|
17 |
MODEL_REPO = "rain1011/pyramid-flow-sd3"
|
18 |
+
MODEL_VARIANT = "diffusion_transformer_384p"
|
19 |
MODEL_DTYPE = "bf16"
|
20 |
|
21 |
def center_crop(image, target_width, target_height):
|
|
|
63 |
# Text-to-video generation function
|
64 |
@spaces.GPU(duration=120)
|
65 |
def generate_video(prompt, image=None, duration=5, guidance_scale=9, video_guidance_scale=5, progress=gr.Progress(track_tqdm=True)):
|
66 |
+
multiplier = 3
|
67 |
+
temp = int(duration * multiplier) + 1 # Convert seconds to temp value (assuming 24 FPS)
|
68 |
torch_dtype = torch.bfloat16 if MODEL_DTYPE == "bf16" else torch.float32
|
69 |
if(image):
|
70 |
+
cropped_image = center_crop(image, 640, 384)
|
71 |
+
resized_image = cropped_image.resize((640, 384))
|
72 |
with torch.no_grad(), torch.cuda.amp.autocast(enabled=True, dtype=torch_dtype):
|
73 |
frames = model.generate_i2v(
|
74 |
prompt=prompt,
|
75 |
input_image=resized_image,
|
76 |
num_inference_steps=[10, 10, 10],
|
77 |
temp=temp,
|
|
|
78 |
video_guidance_scale=video_guidance_scale,
|
79 |
output_type="pil",
|
80 |
save_memory=True,
|
|
|
85 |
prompt=prompt,
|
86 |
num_inference_steps=[20, 20, 20],
|
87 |
video_num_inference_steps=[10, 10, 10],
|
88 |
+
height=384,
|
89 |
+
width=640,
|
90 |
temp=temp,
|
91 |
guidance_scale=guidance_scale,
|
92 |
video_guidance_scale=video_guidance_scale,
|
|
|
94 |
save_memory=True,
|
95 |
)
|
96 |
output_path = f"{str(uuid.uuid4())}_output_video.mp4"
|
97 |
+
export_to_video(frames, output_path, fps=24)
|
98 |
return output_path
|
99 |
|
100 |
# Gradio interface
|
101 |
with gr.Blocks() as demo:
|
102 |
+
gr.Markdown("# Pyramid Flow 384p demo")
|
103 |
+
gr.Markdown("Pyramid Flow is a training-efficient **Autoregressive Video Generation** model based on **Flow Matching**. It is trained only on open-source datasets within 20.7k A100 GPU hours")
|
104 |
+
gr.Markdown("[[Paper](https://arxiv.org/pdf/2410.05954)], [[Model](https://huggingface.co/rain1011/pyramid-flow-sd3)], [[Code](https://github.com/jy0205/Pyramid-Flow)] [[Project Page]](https://pyramid-flow.github.io)")
|
105 |
|
106 |
with gr.Row():
|
107 |
with gr.Column():
|
|
|
109 |
i2v_image = gr.Image(type="pil", label="Input Image")
|
110 |
t2v_prompt = gr.Textbox(label="Prompt")
|
111 |
with gr.Accordion("Advanced settings", open=False):
|
112 |
+
t2v_duration = gr.Slider(minimum=1, maximum=5, value=5, step=1, label="Duration (seconds)", visible=not is_canonical)
|
113 |
+
t2v_guidance_scale = gr.Slider(minimum=1, maximum=15, value=7, step=0.1, label="Guidance Scale")
|
114 |
t2v_video_guidance_scale = gr.Slider(minimum=1, maximum=15, value=5, step=0.1, label="Video Guidance Scale")
|
115 |
t2v_generate_btn = gr.Button("Generate Video")
|
116 |
with gr.Column():
|