version 2

app.py

@@ -1,62 +1,71 @@
 import gradio as gr
 import joblib
 
-# Load models
+# Load models and vectorizer from the models folder
+logistic_model = joblib.load("models/best_logistic_model.pkl")
+svm_model = joblib.load("models/best_svc_model.pkl")
+random_forest_model = joblib.load("models/best_rf_model.pkl")
+knn_model = joblib.load("models/best_knn_model.pkl")
+vectorizer = joblib.load("models/vectorizer.pkl")
+
+# Model selection mapping
 models = {
-    "Logistic Regression": joblib.load("models/best_model.joblib"),
-    "Random Forest": joblib.load("models/random_forest_model.joblib"),
-    "KNN": joblib.load("models/trained_knn_model.joblib"),
+    "Logistic Regression": logistic_model,
+    "SVM": svm_model,
+    "Random Forest": random_forest_model,
+    "KNN": knn_model,
 }
 
-# Load vectorizer
-vectorizer = joblib.load("models/vectorizer.joblib")
-
-# Define prediction function
+# Prediction function
 def predict_sentiment(review, model_name):
-    # Transform the review text using the vectorizer
-    processed_review = vectorizer.transform([review])
-
-    # Select the model
-    model = models[model_name]
-
-    # Make predictions
-    predicted_class = model.predict(processed_review)[0]
-    probabilities = model.predict_proba(processed_review)[0]
-
-    # Define sentiment labels
-    sentiment_labels = ["Negative Comment", "Positive Comment"]
-    predicted_label = sentiment_labels[predicted_class]
-
-    # Return probabilities as percentages
-    positive_percentage = probabilities[1] * 100
-    negative_percentage = probabilities[0] * 100
-
-    return predicted_label, positive_percentage, negative_percentage
-
-# Build Gradio interface
-with gr.Blocks() as interface:
-    gr.Markdown("<h1>Text Classification Models</h1>")
-    gr.Markdown("Choose a model and provide a review to see the sentiment analysis results with probabilities displayed as scales.")
-
-    with gr.Row():
-        with gr.Column():
-            review_input = gr.Textbox(label="Review Comment", placeholder="Type your comment here...")
-            model_selector = gr.Dropdown(
-                choices=list(models.keys()), label="Select Model", value="Logistic Regression"
-            )
-            submit_button = gr.Button("Submit")
-
-        with gr.Column():
-            sentiment_output = gr.Textbox(label="Predicted Sentiment Class", interactive=False)
-            positive_progress = gr.Slider(label="Positive Comment Percentage", minimum=0, maximum=100, interactive=False)
-            negative_progress = gr.Slider(label="Negative Comment Percentage", minimum=0, maximum=100, interactive=False)
-
-    submit_button.click(
-        predict_sentiment,
-        inputs=[review_input, model_selector],
-        outputs=[sentiment_output, positive_progress, negative_progress],
-    )
-
-# Launch the app
-if __name__ == "__main__":
-    interface.launch()
+    try:
+        if not review.strip():
+            return "Error: Review cannot be empty", None
+
+        if model_name not in models:
+            return "Error: Invalid model selected", None
+
+        # Preprocess the text
+        text_vector = vectorizer.transform([review])
+
+        # Predict using the selected model
+        model = models[model_name]
+        prediction = model.predict(text_vector)[0]
+        probabilities = model.predict_proba(text_vector)[0] if hasattr(model, "predict_proba") else None
+
+        # Format the output
+        sentiment = "Positive Feedback" if prediction == 1 else "Negative Feedback"
+        probabilities_output = (
+            {
+                "Positive": probabilities[1],  # Raw probability (0.0 - 1.0)
+                "Negative": probabilities[0],  # Raw probability (0.0 - 1.0)
+            }
+            if probabilities is not None
+            else "Probabilities not available"
+        )
+
+        return sentiment, probabilities_output
+    except Exception as e:
+        # Log the error to the console for debugging
+        print(f"Error in prediction: {e}")
+        return f"Error: {str(e)}", None
+
+# Create Gradio Interface
+inputs = [
+    gr.Textbox(label="Review Comment", placeholder="Enter your review here..."),
+    gr.Dropdown(choices=["Logistic Regression", "SVM", "Random Forest", "KNN"], label="Model"),
+]
+
+outputs = [
+    gr.Textbox(label="Predicted Sentiment Class"),
+    gr.Label(label="Predicted Probability"),
+]
+
+# Launch Gradio App
+gr.Interface(
+    fn=predict_sentiment,
+    inputs=inputs,
+    outputs=outputs,
+    title="Sentiment Analysis",
+    description="Enter a review and select a model to predict sentiment.",
+).launch()

models/best_knn_model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ae5bfda964fad7a1d30fe5cdaaa8e9991670fa81ec5644033b3e3e1d08674e3
+size 105844

models/best_logistic_model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62b2ecaa84982c38c3df38e113605ced250896abf1cb6c762c411a88a6cdeaa1
+size 15215

models/best_rf_model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4d4944dc56c304bda9e45b9b96496e1af61a137ddcae6cc518cc40591514a24
+size 1025289

models/best_svc_model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b2acf2da04be8cb990b9d2fc5573564acdddb7e4b6dfba9c4ff9863193b852f
+size 89531

models/vectorizer.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4df3665f41c3f8e87d1945f3625acb00d8812f7953e10d20f37e6f99d2cf45cd
+size 21760