Pranomvignesh commited on
Commit
510f06a
·
1 Parent(s): c532d5b

Added localization

Browse files
Files changed (3) hide show
  1. .gitignore +1 -0
  2. app.py +15 -7
  3. test.py +0 -14
.gitignore ADDED
@@ -0,0 +1 @@
 
 
1
+ test.py
app.py CHANGED
@@ -10,15 +10,23 @@ model = yolov5.load('./gentle-meadow.pt', device="cpu")
10
 
11
 
12
  def predict(image):
13
- # results = model([image], size=224)
14
- predictions = imageClassifier(image)
15
- print(predictions)
16
 
 
 
 
 
17
  output = {}
18
  for item in predictions:
19
  output[item['label']] = item['score']
20
-
21
- return output
 
 
 
 
 
 
 
22
 
23
 
24
  title = "Detecting Tumors in MRI Images"
@@ -40,8 +48,8 @@ inputs = gr.Image(type="pil", shape=(224, 224),
40
  label="Upload your image for detection")
41
 
42
  outputs = [
43
- # gr.Image(type="pil", label="Tumor Detections"),
44
- gr.Label(label="Tumor Classification")
45
  ]
46
 
47
  interface = gr.Interface(
 
10
 
11
 
12
  def predict(image):
 
 
 
13
 
14
+ predictions = imageClassifier(image)
15
+ maxScore = 0
16
+ predictedLabel = None
17
+ labelsForLocalization = ['meningioma', 'pituitary']
18
  output = {}
19
  for item in predictions:
20
  output[item['label']] = item['score']
21
+ if (maxScore < item['score']):
22
+ maxScore = item['score']
23
+ predictedLabel = item['label']
24
+ if (predictedLabel in labelsForLocalization):
25
+ results = model([image], size=224)
26
+ imageWithLocalization = results.render()[0]
27
+ else:
28
+ imageWithLocalization = image
29
+ return output, imageWithLocalization
30
 
31
 
32
  title = "Detecting Tumors in MRI Images"
 
48
  label="Upload your image for detection")
49
 
50
  outputs = [
51
+ gr.Label(label="Tumor Classification"),
52
+ gr.Image(type="pil", label="Tumor Detections")
53
  ]
54
 
55
  interface = gr.Interface(
test.py DELETED
@@ -1,14 +0,0 @@
1
- from PIL import Image
2
- from gradio_client import Client
3
-
4
- client = Client("PranomVignesh/timri")
5
-
6
- with open('examples/sample_1.jpg', "rb") as f:
7
- image = Image.open(f)
8
- image_bytes = f.read()
9
-
10
- output = client.predict(
11
- image_bytes
12
- )
13
-
14
- print(output)