Update app.py

app.py

@@ -1,134 +1,116 @@
 import streamlit as st
-import pickle
-import nltk
-from nltk.corpus import stopwords
-from nltk.tokenize import word_tokenize
+import joblib
+import re
+import string
 
-# Download NLTK resources
-nltk.download('punkt', quiet=True)
-nltk.download('stopwords', quiet=True)
+# Load the trained model and TF-IDF vectorizer
+knn_model = joblib.load('knn_model.joblib')
+tfidf_vectorizer = joblib.load('tfidf_vectorizer.joblib')
 
-# Load the pre-trained model and vectorizer
-@st.cache_resource
-def load_model():
-    try:
-        # Load TF-IDF Vectorizer
-        with open('tfidf_vectorizer.pkl', 'rb') as vectorizer_file:
-            tfidf_vectorizer = pickle.load(vectorizer_file)
-        
-        # Load KNN Classifier
-        with open('knn_model.pkl', 'rb') as model_file:
-            knn_classifier = pickle.load(model_file)
-        
-        return tfidf_vectorizer, knn_classifier
-    except Exception as e:
-        st.error(f"Error loading model: {e}")
-        return None, None
-
-# Preprocess text function
+# Preprocess the input text (same preprocessing as in the notebook)
 def preprocess_text(text):
-    # Tokenization and preprocessing
-    stop_words = set(stopwords.words('english'))
-    words = word_tokenize(text.lower())
-    words = [word for word in words if word.isalpha() and word not in stop_words]
-    return ' '.join(words)
-
-# Main Streamlit App
-def main():
-    # Set page title and favicon
-    st.set_page_config(
-        page_title="Disease Classification by Symptoms", 
-        page_icon=":medical_symbol:",
-        layout="centered"
-    )
-
-    # Title and description
-    st.title("🩺 Disease Classification Predictor")
-    st.markdown("""
-    ### Predict Potential Diseases Based on Symptoms
-    
-    Enter your symptoms below, and our AI model will help predict possible diseases.
-    """)
-
-    # Load model and vectorizer
-    tfidf_vectorizer, knn_classifier = load_model()
-
-    # Input form for symptoms
-    with st.form(key='symptom_form'):
-        symptoms = st.text_area(
-            "Enter your symptoms:", 
-            placeholder="Example: low appetite, fever, headache",
-            help="Provide a detailed description of your symptoms"
-        )
-        
-        submit_button = st.form_submit_button(label="Predict Disease")
-
-    # Prediction logic
-    if submit_button:
-        if not symptoms:
-            st.warning("Please enter some symptoms.")
-            return
-
-        try:
-            # Preprocess input symptoms
-            preprocessed_symptoms = preprocess_text(symptoms)
-
-            # Transform symptoms using TF-IDF vectorizer
-            symptoms_tfidf = tfidf_vectorizer.transform([preprocessed_symptoms])
-
-            # Predict disease
-            predicted_disease = knn_classifier.predict(symptoms_tfidf)
-
-            # Display prediction
-            st.success(f"Predicted Disease: {predicted_disease[0]}")
-
-            # Additional information (optional)
-            st.info("""
-            ### Disclaimer
-            - This is an AI-based prediction and should not replace professional medical advice
-            - Always consult with a healthcare professional for accurate diagnosis
-            - The prediction is based on machine learning analysis of symptom patterns
-            """)
-
-        except Exception as e:
-            st.error(f"An error occurred during prediction: {e}")
-
-    # Sidebar with additional information
-    st.sidebar.title("About the Model")
-    st.sidebar.markdown("""
-    ### Disease Classification Model
-    - **Algorithm**: K-Nearest Neighbors (KNN)
-    - **Features**: TF-IDF Vectorization
-    - **Trained on**: Symptom to Disease Dataset
+    text = text.lower()  # Convert to lowercase
+    text = re.sub(r'\d+', '', text)  # Remove digits
+    text = text.translate(str.maketrans('', '', string.punctuation))  # Remove punctuation
+    return text
+
+# Prediction function
+def predict_disease(symptom):
+    preprocessed_symptom = preprocess_text(symptom)
+    tfidf_features = tfidf_vectorizer.transform([preprocessed_symptom]).toarray()
+    predicted_disease = knn_model.predict(tfidf_features)
+    return predicted_disease[0]
+
+# Streamlit UI Design
+st.set_page_config(page_title="Disease Prediction App", page_icon="🦠", layout="centered")
+
+# Custom Styling
+st.markdown("""
+    <style>
+    body {
+        background-color: #2E2E2E;
+        color: white;
+        font-family: 'Segoe UI', sans-serif;
+    }
+    .header {
+        font-size: 36px;
+        font-weight: bold;
+        color: #00BFFF;
+        text-align: center;
+        margin-top: 30px;
+        margin-bottom: 15px;
+    }
+    .description {
+        font-size: 16px;
+        color: #dcdcdc;
+        text-align: center;
+        margin-bottom: 20px;
+    }
+    .input-box {
+        background-color: #3E3E3E;
+        border-radius: 8px;
+        padding: 15px;
+        font-size: 16px;
+        font-family: 'Segoe UI', sans-serif;
+        border: none;
+        color: white;
+    }
+    .output {
+        background-color: #5F5F5F;
+        border-radius: 8px;
+        padding: 15px;
+        font-size: 18px;
+        color: #00BFFF;
+        font-weight: bold;
+        text-align: center;
+    }
+    .btn {
+        background-color: #00BFFF;
+        color: white;
+        font-size: 18px;
+        padding: 12px 24px;
+        border-radius: 8px;
+        border: none;
+        cursor: pointer;
+        width: 100%;
+    }
+    .btn:hover {
+        background-color: #008B8B;
+    }
+    footer {
+        margin-top: 40px;
+        text-align: center;
+        font-size: 14px;
+        color: #dcdcdc;
+    }
+    .container {
+        padding: 20px;
+        border-radius: 12px;
+        background-color: #383838;
+        max-width: 500px;
+        margin: auto;
+    }
+    </style>
+""", unsafe_allow_html=True)
+
+# Title and Description
+st.markdown("<div class='header'>🦠 Disease Prediction</div>", unsafe_allow_html=True)
+st.markdown("<div class='description'>Enter your symptoms, and the model will predict the possible disease based on the provided input.</div>", unsafe_allow_html=True)
+
+# Input Box and Prediction Button in a centered container
+with st.container():
+    symptom = st.text_area("Enter symptoms:", height=150, max_chars=500, placeholder="E.g., fever, cough, headache...", key="symptom", label_visibility="collapsed")
     
-    #### How it works:
-    1. Transform symptoms into numerical features
-    2. Compare with known disease patterns
-    3. Predict most likely disease
-    """)
-
-# Run the app
-if __name__ == "__main__":
-    main()
-
-# Additional requirements.txt content
-"""
-streamlit
-scikit-learn
-nltk
-pickle5
-"""
-
-# Deployment Notes
-"""
-Deployment Steps:
-1. Install requirements:
-   pip install -r requirements.txt
-   
-2. Download NLTK resources:
-   python -m nltk.downloader punkt
-   python -m nltk.downloader stopwords
-
-3. Run the Streamlit app:
-   streamlit run app.py
-"""
+    if st.button("Predict", key="predict_button"):
+        if symptom:
+            predicted_disease = predict_disease(symptom)
+            st.markdown(f"<div class='output'>**Predicted Disease: {predicted_disease}**</div>", unsafe_allow_html=True)
+        else:
+            st.warning("Please enter some symptoms to predict the disease.", icon="⚠️")
+
+# Footer with minimalistic text
+st.markdown("""
+    <footer>
+        <p>Powered by AI | Developed for Final Year Project </p>
+    </footer>
+""", unsafe_allow_html=True)