Update app.py

app.py

@@ -1,98 +1,104 @@
 import streamlit as st
 from PIL import Image, ImageOps
 import numpy as np
-import pandas as pd
+import seaborn as sns
 import matplotlib.pyplot as plt
-from keras import tensorflow as tf
+import pandas as pd
+import tensorflow as tf  # Use TensorFlow's Keras API
 
-# Load the TensorFlow model
+# Load the TensorFlow Keras model
 model = tf.keras.models.load_model('gastrointestinal_model.h5', compile=False)
 
 # Load class names
-class_names = ['Normal', 'Ulcerative Colitis', 'Polyp', 'Esophagitis']
+with open('labels.txt', 'r') as f:
+    class_names = f.readlines()
 
 # Function to create plot
-def create_plot(prediction, class_names):
-    df = pd.DataFrame(prediction, index=class_names, columns=['Confidence'])
-    df = df.sort_values(by='Confidence', ascending=False)
-    plt.figure(figsize=(8, 5))
-    plt.bar(df.index, df['Confidence'], color='blue')
-    plt.xlabel('Class')
-    plt.ylabel('Confidence Score')
-    plt.title('Classification Confidence Scores')
-    plt.xticks(rotation=45)
-    plt.tight_layout()
-    return plt
+def create_plot(data):
+    sns.set_theme(style="whitegrid")
+    f, ax = plt.subplots(figsize=(5, 5))
+    sns.set_color_codes("pastel")
+    sns.barplot(x="Total", y="Labels", data=data, label="Total", color="b")
+    sns.set_color_codes("muted")
+    sns.barplot(x="Confidence Score", y="Labels", data=data, label="Confidence Score", color="b")
+    ax.legend(ncol=2, loc="lower right", frameon=True)
+    sns.despine(left=True, bottom=True)
+    return f
 
 # Function to predict gastrointestinal conditions
 def predict_gastrointestinal(img):
+    np.set_printoptions(suppress=True)
+    data = np.ndarray(shape=(1, 224, 224, 3), dtype=np.float32)
     size = (224, 224)
     image_PIL = ImageOps.fit(img, size, Image.LANCZOS)
     image_array = np.asarray(image_PIL)
     normalized_image_array = (image_array.astype(np.float32) / 127.0) - 1
-    data = np.expand_dims(normalized_image_array, axis=0)
+    data[0] = normalized_image_array
+    prediction = model.predict(data)
+    index = np.argmax(prediction)
+    class_name = class_names[index].strip()
+    confidence_score = prediction[0][index]
+    other_class = [name for i, name in enumerate(class_names) if i != index][0].strip()
     
-    prediction = model.predict(data)[0]
-    class_index = np.argmax(prediction)
-    predicted_class = class_names[class_index]
-    confidence_scores = prediction * 100
+    result = {
+        "Labels": [class_name, other_class],
+        "Confidence Score": [confidence_score * 100, (1 - confidence_score) * 100],
+        "Total": 100
+    }
+    data_for_plot = pd.DataFrame.from_dict(result)
+    plot = create_plot(data_for_plot)
     
-    # Create plot
-    plot = create_plot(confidence_scores, class_names)
+    if class_name == "Normal":
+        prediction_text = "The image is classified as Normal."
+    else:
+        prediction_text = f"The image shows signs of {class_name}."
     
-    return predicted_class, plot
+    return prediction_text, plot
 
 # Streamlit app
 st.title("Gastrointestinal Classification Web App")
 
-uploaded_file = st.file_uploader("Upload an image...", type=["jpg", "jpeg", "png"])
+uploaded_file = st.file_uploader("Upload a gastrointestinal image...", type=["jpg", "jpeg", "png"])
 
 if uploaded_file is not None:
     image = Image.open(uploaded_file)
     st.image(image, caption='Uploaded Image', use_column_width=True)
     st.write("Classifying...")
 
-    predicted_class, plot = predict_gastrointestinal(image)
-    st.write(f"Prediction: {predicted_class}")
+    prediction, plot = predict_gastrointestinal(image)
+    st.write(prediction)
     st.pyplot(plot)
 
 # Sample images
 st.markdown("### Sample Images")
 if st.button('Normal Sample'):
-    image = Image.open('normal_sample.jpg')
+    image = Image.open('normal_Sample.jpg')
     st.image(image, caption='Normal Sample Image', use_column_width=True)
     st.write("Classifying...")
-    predicted_class, plot = predict_gastrointestinal(image)
-    st.write(f"Prediction: {predicted_class}")
+    prediction, plot = predict_gastrointestinal(image)
+    st.write(prediction)
     st.pyplot(plot)
 
 if st.button('Ulcerative Colitis Sample'):
     image = Image.open('ulcerative_colitis_sample.jpg')
     st.image(image, caption='Ulcerative Colitis Sample Image', use_column_width=True)
     st.write("Classifying...")
-    predicted_class, plot = predict_gastrointestinal(image)
-    st.write(f"Prediction: {predicted_class}")
+    prediction, plot = predict_gastrointestinal(image)
+    st.write(prediction)
     st.pyplot(plot)
 
 if st.button('Polyp Sample'):
     image = Image.open('polyp_sample.jpg')
     st.image(image, caption='Polyp Sample Image', use_column_width=True)
     st.write("Classifying...")
-    predicted_class, plot = predict_gastrointestinal(image)
-    st.write(f"Prediction: {predicted_class}")
+    prediction, plot = predict_gastrointestinal(image)
+    st.write(prediction)
     st.pyplot(plot)
 
 if st.button('Esophagitis Sample'):
     image = Image.open('esophagitis_sample.jpg')
     st.image(image, caption='Esophagitis Sample Image', use_column_width=True)
     st.write("Classifying...")
-    predicted_class, plot = predict_gastrointestinal(image)
-    st.write(f"Prediction: {predicted_class}")
+    prediction, plot = predict_gastrointestinal(image)
+    st.write(prediction)
     st.pyplot(plot)
-
-# Educational content
-st.markdown("### Learn More About Gastrointestinal Conditions")
-st.markdown("""
-    - [Gastrointestinal Disorders Overview](https://www.mayoclinic.org/diseases-conditions/gastrointestinal-disorders/symptoms-causes/syc-20375441)
-    - [Preventing Gastrointestinal Conditions](https://www.niddk.nih.gov/health-information/digestive-diseases)
-    """)