File size: 2,267 Bytes
82ba409
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
import torch
import torchvision.transforms as transforms
from PIL import Image
import gradio as gr
import warnings
from torchvision import models
import torch.nn as nn

# Define the TransferNet class
class TransferNet(nn.Module):
    def __init__(self, num_classes=2):
        super(TransferNet, self).__init__()
        resnet = models.resnet18(pretrained=True)
        self.features = nn.Sequential(*list(resnet.children())[:-1])
        self.fc = nn.Linear(resnet.fc.in_features, num_classes)

    def forward(self, x):
        x = self.features(x)
        x = x.view(x.size(0), -1)
        x = self.fc(x)
        return x

# Define transformation
transform = transforms.Compose([
    transforms.Resize((224, 224)),
    transforms.RandomHorizontalFlip(p=0.5),
    transforms.RandomVerticalFlip(p=0.5),
    transforms.RandomRotation(30),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
])

class_labels = {0: 'Diabetic Retinopathy', 1: 'No Diabetic Retinopathy'}

# Suppress UserWarnings
warnings.filterwarnings("ignore", category=UserWarning)

def load_model(model_path):
    device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
    model = TransferNet()
    model.load_state_dict(torch.load(model_path, map_location=device))
    model.to(device)
    model.eval()
    return model, device

model_path = "transfer_model.pth"  # Path to your model file
model, device = load_model(model_path)

def predict_image(image):
    img = Image.open(image).convert('RGB')
    img = transform(img).unsqueeze(0).to(device)

    with torch.no_grad():
        output = model(img)
        probabilities = torch.softmax(output, dim=1)
        predicted_class = torch.argmax(probabilities, dim=1).item()

    class_name = class_labels[predicted_class]
    return class_name, probabilities[0].cpu().numpy()

# Create Gradio interface
interface = gr.Interface(
    fn=predict_image,
    inputs=gr.Image(type="filepath"),
    outputs=[
        gr.Textbox(label="Prediction"),
        gr.Textbox(label="Probability of Prediction")
    ],
    title="Diabetic Retinopathy Classification",
    description="Upload an image to classify it as Diabetic Retinopathy or No Diabetic Retinopathy.",
)

interface.launch()