Upload 7 files

CNN_MODEL_0_96414.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9e9e6d4bb4156eb00a716b70b8ac5f304390d3aee5a59aa4747a3db81619a7e
+size 216258760

CNN_MODEL_0_9672.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6f8eeb4eec8096f3aa72de497e7d3681a2dab1fed994aa5edc422f2ba3b9f28
+size 141561592

CNN_MODEL_0_9687.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f26b5ff8e41202202bb8fdcab808437d481f608a5d79d1433e7dc7c697ac1718
+size 237506200

CNN_MODEL_0_9771.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d26361c894bcc2dd650261aff82009a24b105aea9d0d19b6e6ff72b56ae8f22
+size 304564232

CNN_MODEL_0_9794.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c92ab3fad88fc1c7fd4bf5d0c5662b83374fa694e2d42aad2b4dd8497110d4e
+size 854208000

app.py

@@ -0,0 +1,132 @@
+import streamlit as st
+import tensorflow as tf
+from tensorflow import keras
+import cv2
+
+
+st.title(":brain: Brain Tumor Detection (MRI)")
+
+
+
+vggnet_19= keras.models.load_model("CNN_MODEL_0_9687.h5")
+vggnet_16 = keras.models.load_model("CNN_MODEL_0_96414.h5")
+resnet_152 = keras.models.load_model("CNN_MODEL_0_9794.h5")
+mobilenet_v3= keras.models.load_model("CNN_MODEL_0_9771.h5")
+
+
+def check_accuracy(model,img_input):
+    y_pred = model.predict(img_input).argmax(axis=1)
+    prediction = model.predict(img_input)
+    if y_pred[0] == 0:
+        return "Target: glioma_tumor",prediction.max()
+    elif y_pred[0] == 1:
+        return "Target: meningioma_tumor",prediction.max()
+    elif y_pred[0] == 2:
+        return "Target: no_tumor",prediction.max()
+    else:
+        return "Target: pituitary_tumor",prediction.max()
+
+
+
+with st.container():
+    file = st.file_uploader("Upload the MRI Image")
+    if file is not None:
+        print(file)
+        save_image_path = "./upload_images/"+file.name
+        with open(save_image_path,"wb") as f:
+            f.write(file.getbuffer())
+        img = cv2.imread(save_image_path)
+
+col1,col2 = st.columns(2)
+with st.container():
+    if file is not None:
+        with col1:
+            st.write("Original Image:")
+            st.image(file)
+        with col2:
+            st.write("Processed Image:")
+            st.image(cv2.resize(img,(224,224)))
+    else:
+        st.error("First upload image")
+
+with st.container():
+    if file is not None:
+        img = cv2.resize(img,(224,224))
+        img_input = img.reshape((1,224,224,3))
+        vgg19,v19 = check_accuracy(vggnet_19,img_input)
+        vgg16,v16 = check_accuracy(vggnet_16,img_input)
+        resnet,r152 = check_accuracy(resnet_152,img_input)
+        mobilenet,mv3 = check_accuracy(mobilenet_v3,img_input)
+        st.subheader(vgg19+" for"+ " VGGNET-19")
+        st.write("accuracy of the image",v19)
+        st.subheader(vgg16+" for"+ " VGGNET-16")
+        st.write("accuracy of the image",v16)
+        st.subheader(resnet+" for"+ " RESNET-152")
+        st.write("accuracy of the image",r152)
+        st.subheader(mobilenet+" for"+ " MOBILENET-V3")
+        st.write("accuracy of the image",mv3)
+
+
+
+
+
+
+
+
+
+
+
+
+# st.sidebar.title("Choose the Model:")
+# option = st.sidebar.selectbox("Select model",["model1","model2"])
+
+
+# if option == "model1":
+#     st.header("Model 1")
+#     file = st.file_uploader("Upload the MRI Image")
+#     if file is not None:
+#         print(file)
+#         save_image_path = "./upload_images/"+file.name
+#         with open(save_image_path,"wb") as f:
+#             f.write(file.getbuffer())
+#         img = cv2.imread(save_image_path)
+#         img = cv2.resize(img,(250,250))
+#         img_input = img.reshape((1,250,250,3))
+#         y_pred = model1.predict(img_input).argmax(axis=1)
+        
+#         if y_pred[0] == 0:
+#             st.write("Target: glioma_tumor")
+#         elif y_pred[0] == 1:
+#             st.write("Target: meningioma_tumor")
+#         elif y_pred[0] == 2:
+#             st.write("Target: no_tumor")
+#         else:
+#             st.write("Target: pituitary_tumor")
+
+#     else:
+#         st.warning("Not Entered image")
+
+# if option == "model2":
+#     st.header("Model 2")
+#     file = st.file_uploader("Upload the MRI Image")
+#     if file is not None:
+#         print(file)
+#         save_image_path = "./upload_images/"+file.name
+#         with open(save_image_path,"wb") as f:
+#             f.write(file.getbuffer())
+#         img = cv2.imread(save_image_path)
+#         img = cv2.resize(img,(250,250))
+#         img_input = img.reshape((1,250,250,3))
+#         y_pred = model2.predict(img_input).argmax(axis=1)
+        
+#         if y_pred[0] == 0:
+#             st.write("Target: glioma_tumor")
+#         elif y_pred[0] == 1:
+#             st.write("Target: meningioma_tumor")
+#         elif y_pred[0] == 2:
+#             st.write("Target: no_tumor")
+#         else:
+#             st.write("Target: pituitary_tumor")
+
+#     else:
+#         st.warning("Not Entered image")