Spaces:
Runtime error
Runtime error
Upload 4 files
Browse files- app.py +72 -0
- braintumour_model.h5 +3 -0
- labels.txt +4 -0
- requirements.txt +0 -0
app.py
ADDED
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import streamlit as st
|
2 |
+
from keras.models import load_model
|
3 |
+
from PIL import Image, ImageOps
|
4 |
+
import numpy as np
|
5 |
+
# Disable scientific notation for clarity
|
6 |
+
np.set_printoptions(suppress=True)
|
7 |
+
|
8 |
+
# Load the model
|
9 |
+
model = load_model("braintumour_model.h5", compile=False)
|
10 |
+
|
11 |
+
# Load the labels
|
12 |
+
class_names = open("labels.txt", "r").readlines()
|
13 |
+
|
14 |
+
# Function to preprocess the image
|
15 |
+
def preprocess_image(image):
|
16 |
+
# resizing the image to be at least 224x224 and then cropping from the center
|
17 |
+
size = (224, 224)
|
18 |
+
image = ImageOps.fit(image, size, Image.Resampling.LANCZOS)
|
19 |
+
|
20 |
+
# turn the image into a numpy array
|
21 |
+
image_array = np.asarray(image)
|
22 |
+
|
23 |
+
# Normalize the image
|
24 |
+
normalized_image_array = (image_array.astype(np.float32) / 127.5) - 1
|
25 |
+
|
26 |
+
# Return the preprocessed image
|
27 |
+
return normalized_image_array
|
28 |
+
|
29 |
+
# Function to predict the image class
|
30 |
+
def predict_image_class(image):
|
31 |
+
# Create the array of the right shape to feed into the keras model
|
32 |
+
data = np.ndarray(shape=(1, 224, 224, 3), dtype=np.float32)
|
33 |
+
|
34 |
+
# Load the image into the array
|
35 |
+
data[0] = image
|
36 |
+
|
37 |
+
# Predict the model
|
38 |
+
prediction = model.predict(data)
|
39 |
+
index = np.argmax(prediction)
|
40 |
+
class_name = class_names[index]
|
41 |
+
confidence_score = prediction[0][index]
|
42 |
+
|
43 |
+
return class_name[2:], round(confidence_score, 2)*100
|
44 |
+
|
45 |
+
# Streamlit app
|
46 |
+
def main():
|
47 |
+
# Set page title and icon
|
48 |
+
st.set_page_config(page_title="Brain Tumour Classifier", page_icon="🧠")
|
49 |
+
|
50 |
+
st.title("Brain Tumour Classifier")
|
51 |
+
|
52 |
+
# Upload image file
|
53 |
+
uploaded_file = st.file_uploader("Upload the MRI", type=["jpg", "jpeg", "png"])
|
54 |
+
|
55 |
+
if uploaded_file is not None:
|
56 |
+
image = Image.open(uploaded_file).convert("RGB")
|
57 |
+
st.image(image, caption="Uploaded Image", use_column_width=True)
|
58 |
+
|
59 |
+
preprocessed_image = preprocess_image(image)
|
60 |
+
|
61 |
+
# Predict the image class
|
62 |
+
class_name, confidence_score = predict_image_class(preprocessed_image)
|
63 |
+
|
64 |
+
# Display the prediction
|
65 |
+
st.subheader("RESULT")
|
66 |
+
st.write("Class:", "**" + class_name + "**")
|
67 |
+
st.write("Prediction Probability:", "**" + str(confidence_score) + "%**")
|
68 |
+
|
69 |
+
|
70 |
+
# Run the app
|
71 |
+
if __name__ == "__main__":
|
72 |
+
main()
|
braintumour_model.h5
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:378e3392f9db5be6a8055a7a691a670c28275b5d09f6402d2b1b59ee3c205c37
|
3 |
+
size 2453432
|
labels.txt
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
0 Glioma Tumour
|
2 |
+
1 Meningioma Tumour
|
3 |
+
2 No Tumor
|
4 |
+
3 Pituitary Tumour
|
requirements.txt
ADDED
Binary file (5.83 kB). View file
|
|