Project Files added

app.py

@@ -0,0 +1,37 @@
+from PIL import Image
+import tensorflow as tf
+import streamlit as st
+from deep_learning_pipeline import PredictionPipeline
+
+st.title('Malaria Infected Cell Detection using X-ray Images')
+st.write('This Project is built using CNN (Convolutional Neural Networks) Transfer Learning model that helps to predict whether the given X-ray image of the cell is Malaria Infected or Healthy!!')
+
+st.write('')
+st.write('')
+
+
+uploaded_file = st.file_uploader("Upload an image", type=["png", "jpg", "jpeg"])
+
+if uploaded_file is not None:
+    # Process the uploaded image here
+    with st.container():
+        col1, col2 = st.columns([3, 2])
+        col1.image(uploaded_file, caption="Uploaded Image", use_column_width=True)
+
+        if st.button('Predict!!'):
+            pipeline = PredictionPipeline()
+            resnet152v2_y_pred, resnet152v2_y_probs = pipeline.predict(input_img=uploaded_file)
+            col2.balloons()
+            if resnet152v2_y_pred[0][0] == 1:
+                col2.subheader('ResNET 152V2 model: ')
+                col2.success(f'{pipeline.CLASS_NAMES[1]}')
+                r_acc = '{:.2f}'.format(100*(resnet152v2_y_probs[0][0]))
+                col2.success(f'Accuracy: {r_acc}%')
+            elif resnet152v2_y_pred[0][0] == 0:
+                col2.subheader('ResNET 152V2 model: ')
+                col2.success(f'{pipeline.CLASS_NAMES[0]}')
+                r_acc = '{:.2f}'.format(100*(1-resnet152v2_y_probs[0][0]))
+                col2.success(f'Accuracy: {r_acc}%')
+
+            elif resnet152v2_y_pred[[0]] == -1:
+                col2.error('Error!! Model needs shape (224, 224, 3), but your image is of shape (224, 224,4)')

deep_learning_pipeline.py

@@ -0,0 +1,26 @@
+import tensorflow as tf
+from tensorflow import keras
+from keras.models import load_model
+from PIL import Image
+
+
+class PredictionPipeline():
+    def __init__(self) -> None:
+        self.CLASS_NAMES = ['Malaria Infected cell', 'Healthy Cell']
+        self.IMG_SIZE = 224
+
+    def predict(self, input_img):
+        # Loading ResNet152v2 model
+        resnet_152v2_model = load_model('model_resnet152v2.h5')
+        # Image Preprocessing
+        image = Image.open(input_img)
+        image = tf.cast(image, dtype=tf.float32)
+        image = image / 255.0
+        input_tensor = tf.expand_dims(tf.image.resize(image, [self.IMG_SIZE, self.IMG_SIZE]), axis=0)
+        # Making Predictions
+        try:
+            resnet_152v2_y_probs = resnet_152v2_model.predict(input_tensor)
+        except ValueError as err:
+            return [[-1]], err, err, err
+        else:
+            return tf.round(resnet_152v2_y_probs), resnet_152v2_y_probs

model_resnet152v2.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2450ac6a99a8355c4d93b8334a88d9ee46ea260572f771a70115c00a5cd59e58
+size 542988016

requirements.txt

@@ -0,0 +1,3 @@
+tensorflow==2.13.0
+Pillow==10.1.0
+streamlit