Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -5,7 +5,6 @@ from PIL import Image
|
|
5 |
import numpy as np
|
6 |
from torchvision import transforms
|
7 |
|
8 |
-
# Set up device
|
9 |
device = "cuda" if torch.cuda.is_available() else "cpu"
|
10 |
|
11 |
# Initialize the inpainting model
|
@@ -18,65 +17,38 @@ except Exception as e:
|
|
18 |
# Load a test image
|
19 |
def load_test_image():
|
20 |
try:
|
21 |
-
|
|
|
22 |
except Exception as e:
|
23 |
print(f"Error loading test image: {e}")
|
24 |
return None
|
25 |
|
26 |
def process_image(prompt, image, style, upscale_factor, inpaint):
|
27 |
-
|
28 |
-
|
29 |
if image is None:
|
30 |
-
image
|
31 |
-
if image is None:
|
32 |
-
return None, "No image received and failed to load test image."
|
33 |
|
|
|
34 |
print(f"Received image type: {type(image)}")
|
35 |
-
|
36 |
if isinstance(image, np.ndarray):
|
37 |
-
print(f"Image shape: {image.shape}")
|
38 |
image = Image.fromarray(image)
|
39 |
elif isinstance(image, torch.Tensor):
|
40 |
-
print(f"Image tensor shape: {image.shape}")
|
41 |
image = transforms.ToPILImage()(image)
|
42 |
-
elif isinstance(image, Image.Image):
|
43 |
-
print("Image is already in PIL format.")
|
44 |
-
else:
|
45 |
return None, f"Unsupported image format: {type(image)}."
|
46 |
-
|
47 |
-
|
48 |
-
if not isinstance(image, Image.Image):
|
49 |
-
return None, "Error: Image format conversion failed."
|
50 |
-
|
51 |
-
# Log the input parameters
|
52 |
-
print(f"Prompt: {prompt}")
|
53 |
-
print(f"Style: {style}")
|
54 |
-
print(f"Upscale Factor: {upscale_factor}")
|
55 |
-
print(f"Inpaint: {inpaint}")
|
56 |
-
|
57 |
-
# Example placeholder logic for using the pipeline
|
58 |
-
if inpaint and inpaint_model:
|
59 |
-
result = inpaint_model(prompt=prompt, image=image, guidance_scale=7.5)
|
60 |
-
else:
|
61 |
-
result = inpaint_model(prompt=prompt, guidance_scale=7.5) if inpaint_model else None
|
62 |
-
|
63 |
-
# Check if the result is valid
|
64 |
-
if result and hasattr(result, 'images') and len(result.images) > 0:
|
65 |
-
return result.images[0], None # Return image and no error
|
66 |
-
else:
|
67 |
-
return None, "Error: No image returned from model." # Return no image and an error message
|
68 |
-
|
69 |
except Exception as e:
|
70 |
error_message = f"Error in process_image function: {e}"
|
71 |
print(error_message)
|
72 |
-
return None, error_message
|
73 |
|
74 |
-
# Define the Gradio interface
|
75 |
with gr.Blocks() as demo:
|
76 |
with gr.Row():
|
77 |
with gr.Column():
|
78 |
prompt_input = gr.Textbox(label="Enter your prompt")
|
79 |
-
image_input = gr.Image(label="Image (for inpainting)", type="pil")
|
80 |
style_input = gr.Dropdown(choices=["Fooocus Style", "SAI Anime"], label="Select Style")
|
81 |
upscale_input = gr.Slider(minimum=1, maximum=4, step=1, label="Upscale Factor")
|
82 |
inpaint_input = gr.Checkbox(label="Enable Inpainting")
|
@@ -88,8 +60,7 @@ with gr.Blocks() as demo:
|
|
88 |
generate_button.click(
|
89 |
process_image,
|
90 |
inputs=[prompt_input, image_input, style_input, upscale_input, inpaint_input],
|
91 |
-
outputs=[output_image, error_output]
|
92 |
)
|
93 |
|
94 |
-
# Launch the interface
|
95 |
demo.launch()
|
|
|
5 |
import numpy as np
|
6 |
from torchvision import transforms
|
7 |
|
|
|
8 |
device = "cuda" if torch.cuda.is_available() else "cpu"
|
9 |
|
10 |
# Initialize the inpainting model
|
|
|
17 |
# Load a test image
|
18 |
def load_test_image():
|
19 |
try:
|
20 |
+
# Provide the absolute path to a test image
|
21 |
+
return Image.open("/absolute/path/to/your/test_image.png")
|
22 |
except Exception as e:
|
23 |
print(f"Error loading test image: {e}")
|
24 |
return None
|
25 |
|
26 |
def process_image(prompt, image, style, upscale_factor, inpaint):
|
27 |
+
if image is None:
|
28 |
+
image = load_test_image()
|
29 |
if image is None:
|
30 |
+
return None, "No image received and failed to load test image."
|
|
|
|
|
31 |
|
32 |
+
try:
|
33 |
print(f"Received image type: {type(image)}")
|
|
|
34 |
if isinstance(image, np.ndarray):
|
|
|
35 |
image = Image.fromarray(image)
|
36 |
elif isinstance(image, torch.Tensor):
|
|
|
37 |
image = transforms.ToPILImage()(image)
|
38 |
+
elif not isinstance(image, Image.Image):
|
|
|
|
|
39 |
return None, f"Unsupported image format: {type(image)}."
|
40 |
+
|
41 |
+
return image, None
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
42 |
except Exception as e:
|
43 |
error_message = f"Error in process_image function: {e}"
|
44 |
print(error_message)
|
45 |
+
return None, error_message
|
46 |
|
|
|
47 |
with gr.Blocks() as demo:
|
48 |
with gr.Row():
|
49 |
with gr.Column():
|
50 |
prompt_input = gr.Textbox(label="Enter your prompt")
|
51 |
+
image_input = gr.Image(label="Image (for inpainting)", type="pil")
|
52 |
style_input = gr.Dropdown(choices=["Fooocus Style", "SAI Anime"], label="Select Style")
|
53 |
upscale_input = gr.Slider(minimum=1, maximum=4, step=1, label="Upscale Factor")
|
54 |
inpaint_input = gr.Checkbox(label="Enable Inpainting")
|
|
|
60 |
generate_button.click(
|
61 |
process_image,
|
62 |
inputs=[prompt_input, image_input, style_input, upscale_input, inpaint_input],
|
63 |
+
outputs=[output_image, error_output]
|
64 |
)
|
65 |
|
|
|
66 |
demo.launch()
|