Update app.py

app.py

@@ -1,14 +1,17 @@
-import gradio as gr
 import pandas as pd
 import numpy as np
 import zipfile
 import os
-from sklearn.preprocessing import StandardScaler
+import gradio as gr
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import LabelEncoder, StandardScaler
 from sklearn.ensemble import RandomForestClassifier
+from sklearn.metrics import accuracy_score, classification_report
+from imblearn.over_sampling import SMOTE
 
 # Extract ZIP file
-zip_file_path = "LUNG_CANCER.zip"
-extract_folder = "LUNG_CANCER_DATA"
+zip_file_path = "LUNG CANCER.zip"
+extract_folder = "./LUNG_CANCER_DATA"
 with zipfile.ZipFile(zip_file_path, 'r') as zip_ref:
     zip_ref.extractall(extract_folder)
 
@@ -23,79 +26,52 @@ df['LUNG_CANCER'] = df['LUNG_CANCER'].map({'YES': 1, 'NO': 0})
 # Splitting dataset
 X = df.drop(columns=['LUNG_CANCER'])
 y = df['LUNG_CANCER']
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42, stratify=y)
+
+# Handling class imbalance
+smote = SMOTE(random_state=42)
+X_train_resampled, y_train_resampled = smote.fit_resample(X_train, y_train)
 
 # Scaling features
 scaler = StandardScaler()
-X_scaled = scaler.fit_transform(X)
+X_train_resampled = scaler.fit_transform(X_train_resampled)
+X_test = scaler.transform(X_test)
 
-# Train model
-model = RandomForestClassifier(n_estimators=100, random_state=42)
-model.fit(X_scaled, y)
+# Model training
+model = RandomForestClassifier(n_estimators=200, random_state=42)
+model.fit(X_train_resampled, y_train_resampled)
 
-# Prediction function
-def predict_lung_cancer(GENDER, AGE, SMOKING, YELLOW_FINGERS, ANXIETY, PEER_PRESSURE, CHRONIC_DISEASE, 
-                         FATIGUE, ALLERGY, WHEEZING, ALCOHOL_CONSUMING, COUGHING, SHORTNESS_OF_BREATH, 
-                         SWALLOWING_DIFFICULTY, CHEST_PAIN):
-    
-    # Gender Encoding: Male -> 0, Female -> 1
-    gender_value = 0 if GENDER.lower() == "male" else 1
-    
-    # Convert "Yes" -> 1, "No" -> 0
-    feature_values = [gender_value, AGE] + [
-        1 if x.lower() == "yes" else 0 for x in [
-            SMOKING, YELLOW_FINGERS, ANXIETY, PEER_PRESSURE, CHRONIC_DISEASE, 
-            FATIGUE, ALLERGY, WHEEZING, ALCOHOL_CONSUMING, COUGHING, SHORTNESS_OF_BREATH, 
-            SWALLOWING_DIFFICULTY, CHEST_PAIN
-        ]
-    ]
-    
-    # Convert to NumPy array and Reshape
-    features = np.array([feature_values])
-    
-    # **Apply the same scaling transformation used during training**
-    features_scaled = scaler.transform(features)
-    
-    # **Predict using the trained model**
-    prediction = model.predict(features_scaled)
-    
-    return "Lung Cancer Detected" if prediction[0] == 1 else "No Lung Cancer"
-# Create GUI with Gradio
-with gr.Blocks() as demo:
-    gr.Markdown("# 🩺 Lung Cancer Detection AI")
-    gr.Markdown("Provide patient details and check if lung cancer is detected.")
-    
-    with gr.Row():
-        gender = gr.Radio(["Male", "Female"], label="Gender")
-        age = gr.Slider(20, 90, step=1, label="Age")
-
-    with gr.Row():
-        smoking = gr.Radio(["No", "Yes"], label="Smoking")
-        yellow_fingers = gr.Radio(["No", "Yes"], label="Yellow Fingers")
-        anxiety = gr.Radio(["No", "Yes"], label="Anxiety")
-        peer_pressure = gr.Radio(["No", "Yes"], label="Peer Pressure")
-        chronic_disease = gr.Radio(["No", "Yes"], label="Chronic Disease")
-    
-    with gr.Row():
-        fatigue = gr.Radio(["No", "Yes"], label="Fatigue")
-        allergy = gr.Radio(["No", "Yes"], label="Allergy")
-        wheezing = gr.Radio(["No", "Yes"], label="Wheezing")
-        alcohol_consuming = gr.Radio(["No", "Yes"], label="Alcohol Consuming")
-        coughing = gr.Radio(["No", "Yes"], label="Coughing")
+# Model evaluation
+y_pred = model.predict(X_test)
+accuracy = accuracy_score(y_test, y_pred)
+print(f"Model Accuracy: {accuracy:.2f}")
+print("Classification Report:\n", classification_report(y_test, y_pred))
 
-    with gr.Row():
-        shortness_of_breath = gr.Radio(["No", "Yes"], label="Shortness of Breath")
-        swallowing_difficulty = gr.Radio(["No", "Yes"], label="Swallowing Difficulty")
-        chest_pain = gr.Radio(["No", "Yes"], label="Chest Pain")
-    
-    predict_button = gr.Button("Predict")
-    output_text = gr.Textbox(label="Prediction Result")
+# Gradio Prediction Function
+def predict_lung_cancer(*features):
+    features = np.array(features).reshape(1, -1)
+    features = scaler.transform(features)
+    prediction = model.predict(features)
+    return "Lung Cancer Detected" if prediction[0] == 1 else "No Lung Cancer"
 
-    predict_button.click(
-        predict_lung_cancer, 
-        inputs=[gender, age, smoking, yellow_fingers, anxiety, peer_pressure, chronic_disease, 
-                fatigue, allergy, wheezing, alcohol_consuming, coughing, shortness_of_breath, 
-                swallowing_difficulty, chest_pain], 
-        outputs=output_text
-    )
+# Gradio Interface
+inputs = [
+    gr.Number(label="Gender (0: Male, 1: Female)"),
+    gr.Number(label="Age"),
+    gr.Number(label="Smoking"),
+    gr.Number(label="Yellow Fingers"),
+    gr.Number(label="Anxiety"),
+    gr.Number(label="Peer Pressure"),
+    gr.Number(label="Chronic Disease"),
+    gr.Number(label="Fatigue"),
+    gr.Number(label="Allergy"),
+    gr.Number(label="Wheezing"),
+    gr.Number(label="Alcohol Consuming"),
+    gr.Number(label="Coughing"),
+    gr.Number(label="Shortness of Breath"),
+    gr.Number(label="Swallowing Difficulty"),
+    gr.Number(label="Chest Pain")
+]
 
+demo = gr.Interface(fn=predict_lung_cancer, inputs=inputs, outputs="text", title="Lung Cancer Prediction")
 demo.launch()