Update app.py

app.py

@@ -1,39 +1,22 @@
 import gradio as gr
+import joblib
 import numpy as np
-from joblib import load
 
-# Step 6.1: Load your trained models (ensure these .pkl files are correctly uploaded and paths are correct)
-logreg_model = load('best_lr_model.pkl')  # Update with your correct path
-knn_model = load('best_knn_model.pkl')  # Update with your correct path
-svc_model = load('best_svc_model.pkl')  # Update with your correct path
-rf_model = load('best_rf_model.pkl')  # Update with your correct path
+# Load models (change paths to where your .pkl files are stored)
+logreg_model = joblib.load('best_lr_model.pkl')
+knn_model = joblib.load('best_knn_model.pkl')
+svc_model = joblib.load('best_svc_model.pkl')
+rf_model = joblib.load('best_rf_model.pkl')
 
-# Step 6.2: Define different input components
+# Preprocessing: Load the same vectorizer you used during training
+from sklearn.feature_extraction.text import CountVectorizer
+vectorizer = joblib.load('vectorizer.pkl')
 
-# a. define textbox input for review comment (as per your example)
-review_input = gr.Textbox(label="Review Comment", lines=2, placeholder="Enter your review comment here...")
-
-# b. define dropdown input for model selection (mimicking the dropdown for models like Logistic Regression, K-Nearest Neighbors, etc.)
-model_dropdown = gr.Dropdown(
-    choices=["Logistic Regression", "K-Nearest Neighbors", "Random Forest", "SVM"], 
-    label="Select Model"
-)
-
-# Step 6.3: Define output component for predicted sentiment class and probability
-output_sentiment = gr.Textbox(label="Predicted Sentiment Class", placeholder="Predicted sentiment will appear here...")
-output_probability = gr.Textbox(label="Predicted Probability", placeholder="Probabilities will appear here...")
-
-# Step 6.4: Define a function to predict sentiment using the selected model
+# Define the function to predict sentiment
 def predict_sentiment(review_text, model_name):
-    # Preprocess the review_text using vectorizer (replace with your preprocessing logic)
-    # For example, use TfidfVectorizer or CountVectorizer as needed
-    # Here I'm assuming you've preprocessed the input text in the same way you did during training
+    # Preprocess the review text using the same vectorizer used during training
+    transformed_text = vectorizer.transform([review_text])  # Transform the input review text
 
-    # Dummy text transformation for demonstration purposes
-    # Replace with actual vectorizer code (e.g., vectorizer.transform([review_text])) if needed
-    transformed_text = np.array([review_text])  # Example transformation
-
-    # Choose the model based on the selected dropdown
     if model_name == "Logistic Regression":
         model = logreg_model
     elif model_name == "K-Nearest Neighbors":
@@ -43,19 +26,31 @@ def predict_sentiment(review_text, model_name):
     elif model_name == "SVM":
         model = svc_model
 
-    # Get the model prediction and probabilities
+    # Make predictions and get probabilities
     prob = model.predict_proba(transformed_text)
     sentiment = model.predict(transformed_text)[0]
-
-    # Return the predicted sentiment and the probabilities
-    return sentiment, f"Positive: {prob[0][1]*100:.2f}%, Negative: {prob[0][0]*100:.2f}%"
-
-# Step 6.5: Put all components together in Gradio's interface
-gr.Interface(
-    fn=predict_sentiment,
-    inputs=[review_input, model_dropdown],
-    outputs=[output_sentiment, output_probability],
-    live=True,  # Enable live updates as the user types
-    description="Sentiment Analysis Model",
-    theme="compact"  # Use a more compact layout for the interface
-).launch()
+    
+    # Return the predicted sentiment class and probabilities
+    sentiment_class = "Positive" if sentiment == 1 else "Negative"
+    probability = f"Positive Comment: {prob[0][1]*100:.2f}%, Negative Comment: {prob[0][0]*100:.2f}%"
+    
+    return sentiment_class, probability
+
+# Step 6.4: Put all components together in Gradio's interface
+with gr.Blocks() as demo:
+    # Input components
+    review_input = gr.Textbox(label="Review Comment", lines=2, placeholder="Enter your review comment here...")
+    model_dropdown = gr.Dropdown(choices=["Logistic Regression", "K-Nearest Neighbors", "Random Forest", "SVM"], label="Select Model")
+    
+    # Output components
+    output_sentiment = gr.Textbox(label="Predicted Sentiment Class", placeholder="Predicted sentiment will appear here...")
+    output_probability = gr.Textbox(label="Predicted Probability", placeholder="Probabilities will appear here...")
+    
+    # Submit Button
+    submit_button = gr.Button("Submit")
+    
+    # Define the action when the button is clicked
+    submit_button.click(fn=predict_sentiment, inputs=[review_input, model_dropdown], outputs=[output_sentiment, output_probability])
+
+# Launch the interface
+demo.launch()