Upload 3 files
Browse files- app.py +78 -0
- best.pt +3 -0
- requirements.txt +3 -0
app.py
ADDED
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
import torch
|
3 |
+
from ultralyticsplus import YOLO, render_result
|
4 |
+
from PIL import Image
|
5 |
+
import os
|
6 |
+
|
7 |
+
def yolov8_func(image,
|
8 |
+
image_size,
|
9 |
+
conf_thresold=0.4,
|
10 |
+
iou_thresold=0.50):
|
11 |
+
|
12 |
+
# Load the YOLOv8 model
|
13 |
+
model_path = "best.pt"
|
14 |
+
model = YOLO(model_path) # Use your custom model path here
|
15 |
+
|
16 |
+
# Make predictions
|
17 |
+
result = model.predict(image, conf=conf_thresold, iou=iou_thresold, imgsz=image_size)
|
18 |
+
|
19 |
+
# Access object detection results
|
20 |
+
boxes = result[0].boxes # Bounding boxes
|
21 |
+
num_boxes = len(boxes) # Count the number of bounding boxes (detections)
|
22 |
+
|
23 |
+
# Print object detection details (optional)
|
24 |
+
print("Object type: ", boxes.cls)
|
25 |
+
print("Confidence: ", boxes.conf)
|
26 |
+
print("Coordinates: ", boxes.xyxy)
|
27 |
+
print(f"Number of bounding boxes: {num_boxes}")
|
28 |
+
|
29 |
+
# Categorize based on number of boxes (detections) and provide recommendations
|
30 |
+
if num_boxes > 10:
|
31 |
+
severity = "Worse"
|
32 |
+
recommendation = "It is recommended to see a dermatologist and start stronger acne treatment."
|
33 |
+
elif 5 <= num_boxes <= 10:
|
34 |
+
severity = "Medium"
|
35 |
+
recommendation = "You should follow a consistent skincare routine with proper cleansing and moisturizing."
|
36 |
+
else:
|
37 |
+
severity = "Good"
|
38 |
+
recommendation = "Your skin looks good! Keep up with your current skincare routine."
|
39 |
+
|
40 |
+
print(f"Acne condition: {severity}")
|
41 |
+
print(f"Recommendation: {recommendation}")
|
42 |
+
|
43 |
+
# Render the result (with bounding boxes/labels)
|
44 |
+
render = render_result(model=model, image=image, result=result[0])
|
45 |
+
|
46 |
+
# Save the rendered image (with predictions)
|
47 |
+
predicted_image_save_path = "predicted_image.jpg"
|
48 |
+
render.save(predicted_image_save_path)
|
49 |
+
|
50 |
+
# Return the saved image, severity, and recommendation for Gradio output
|
51 |
+
return predicted_image_save_path, f"Acne condition: {severity}", recommendation
|
52 |
+
|
53 |
+
# Define inputs for the Gradio app
|
54 |
+
inputs = [
|
55 |
+
gr.Image(type="filepath", label="Input Image"),
|
56 |
+
gr.Slider(minimum=320, maximum=1280, step=32, value=640, label="Image Size"),
|
57 |
+
gr.Slider(minimum=0, maximum=1, step=0.05, value=0.15, label="Confidence Threshold"),
|
58 |
+
gr.Slider(minimum=0, maximum=1, step=0.05, value=0.15, label="IOU Threshold")
|
59 |
+
]
|
60 |
+
|
61 |
+
# Define the output for the Gradio app (image + text for severity and recommendation)
|
62 |
+
outputs = [
|
63 |
+
gr.Image(type="filepath", label="Output Image"),
|
64 |
+
gr.Textbox(label="Acne Condition"),
|
65 |
+
gr.Textbox(label="Recommendation")
|
66 |
+
]
|
67 |
+
|
68 |
+
# Set the title of the Gradio app
|
69 |
+
title = "YOLOv8: An Object Detection for Acne"
|
70 |
+
|
71 |
+
# Create the Gradio interface
|
72 |
+
yolo_app = gr.Interface(fn=yolov8_func,
|
73 |
+
inputs=inputs,
|
74 |
+
outputs=outputs,
|
75 |
+
title=title)
|
76 |
+
|
77 |
+
# Launch the app
|
78 |
+
yolo_app.launch(debug=True)
|
best.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:88dd8766ff1f53969339c030cb78f8736148c25a1bbeec111d171e5126ce08e4
|
3 |
+
size 19956335
|
requirements.txt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
gradio==5.0.0
|
2 |
+
torch
|
3 |
+
ultralyticsplus==0.1.0
|