File size: 2,427 Bytes
92aa748
 
a2a5b20
92aa748
 
d444637
5950f80
8b7fd62
5950f80
e8bfb0c
5950f80
c0f4c8a
5950f80
c0f4c8a
 
 
 
 
 
 
5950f80
 
 
f404c6c
415fd74
dc87457
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8b7fd62
d444637
3b82920
42e38b7
741c74f
f825554
 
 
 
96f2bb3
f825554
1c07639
 
 
 
 
 
 
 
 
d444637
92aa748
 
d444637
b8f132d
187b87b
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
import gradio as gr
from fastai.vision.all import *
from efficientnet_pytorch import EfficientNet 


title = "COVID_19 Infection Detectation App!"
head = (  
    "<body>"
        "<center>"
            "<img src='/file=Gradcam.png' width=200>"
            "<h2>"
                "This Space demonstrates a model based on efficientnet base model. The Model was trained to classify chest xray image. To test it, "
            "</h2>"
            "<h3>"
                "Use the Example Images Provided or Upload your own xray images the space provided." 
            "</h3>"
            "<h3>"
                "!!!PLEASE NOTE MODEL WAS TRAINED and VALIDATED USING PNG FILES!!!" 
            "</h3>"
    "</center>"
            "<p>"
              "The model is trained using [anasmohammedtahir/covidqu(https://www.kaggle.com/datasets/anasmohammedtahir/covidqu) dataset"
            "</p>"
    
            "<p>"
                "The researchers of Qatar University have compiled the COVID-QU-Ex dataset, which consists of 33,920 chest X-ray (CXR) images including:"
            "</p>"
            "<ul>"
              "<li>" 
                    "11,956 COVID-19"
              "</li>"    
              "<li>" 
                    "11,263 Non-COVID infections (Viral or Bacterial Pneumonia)"
              "</li>"     
              "<li>" 
                    "10,701 Normal"
              "</li>"     
            "</ul>"
            "<p>"
                "Ground-truth lung segmentation masks are provided for the entire dataset. This is the largest ever created lung mask dataset."
            "</p>"
    "</body>"
)
description = head

examples = [
    ['covid/covid_1038.png'], ['covid/covid_1034.png'], 
    ['covid/cd.png'], ['covid/covid_1021.png'], 
    ['covid/covid_1027.png'], ['covid/covid_1042.png'], 
    ['covid/covid_1031.png']
]

#learn = load_learner('model/predictcovidfastaifinal18102023.pkl')
learn = load_learner('model/final_20102023_eb7_model.pkl')

categories = learn.dls.vocab

def predict_image(get_image):
   pred, idx, probs = learn.predict(get_image)
   return dict(zip(categories, map(float, probs)))

interpretation="default"
enable_queue=True

gr.Interface(fn=predict_image, inputs=gr.Image(shape=(224,224)),
             outputs = gr.Label(num_top_classes=3),title=title,description=description,examples=examples, interpretation=interpretation,enable_queue=enable_queue).launch(share=False)