Update app.py

app.py

@@ -3,63 +3,89 @@ import pandas as pd
 import json
 import joblib
 import torch
+import os
 
-# Function to load model safely on CPU
-def load_model(model_path):
-    """Loads the model, ensuring it is mapped to the CPU."""
-    return torch.load(model_path, map_location=torch.device('cpu'))
+# Function to load models safely
+def load_model(model_path, is_pytorch=False):
+    """Loads a model based on the file type, ensuring safe execution."""
+    try:
+        if is_pytorch:
+            return torch.load(model_path, map_location=torch.device('cpu'), weights_only=False)
+        else:
+            return joblib.load(model_path)
+    except Exception as e:
+        st.error(f"Error loading model {model_path}: {e}")
+        return None
 
-# Model file paths (ensure they are uploaded in the correct directory)
-distilbert_model_path = "models/distilbert_model.joblib"
+# Model file paths (ensure they exist)
+distilbert_model_path = "models/distilbert_model.pt"
 bert_topic_model_path = "models/bertopic_model.joblib"
 recommendation_model_path = "models/recommendation_model.joblib"
 
-# Load all models with CPU mapping
-distilbert_model = load_model(distilbert_model_path)
+# Check if models exist before loading
+if not os.path.exists(distilbert_model_path):
+    st.error(f"Model not found: {distilbert_model_path}")
+if not os.path.exists(bert_topic_model_path):
+    st.error(f"Model not found: {bert_topic_model_path}")
+if not os.path.exists(recommendation_model_path):
+    st.error(f"Model not found: {recommendation_model_path}")
+
+# Load models safely
+distilbert_model = load_model(distilbert_model_path, is_pytorch=True)
 bert_topic_model = load_model(bert_topic_model_path)
 recommendation_model = load_model(recommendation_model_path)
 
 # Streamlit app layout
-st.title("Intelligent Customer Feedback Analyzer")
+st.title("📊 Intelligent Customer Feedback Analyzer")
 st.write("Analyze customer feedback for sentiment, topics, and get personalized recommendations.")
 
 # User input for customer feedback file
-uploaded_file = st.file_uploader("Upload a Feedback File (CSV, JSON, TXT)", type=["csv", "json", "txt"])
+uploaded_file = st.file_uploader("📂 Upload a Feedback File (CSV, JSON, TXT)", type=["csv", "json", "txt"])
 
 # Function to extract feedback text from different file formats
 def extract_feedback(file):
-    if file.type == "text/csv":
-        df = pd.read_csv(file)
-        feedback_text = []
-        for column in df.columns:
-            feedback_text.extend(df[column].dropna().astype(str).tolist())
-        return feedback_text
-    elif file.type == "application/json":
-        json_data = json.load(file)
-        feedback_text = [item.get('feedback', '') for item in json_data if isinstance(item, dict)]
-        return feedback_text
-    elif file.type == "text/plain":
-        return [file.getvalue().decode("utf-8")]
-    else:
-        return ["Unsupported file type"]
+    """Extracts text data from CSV, JSON, or TXT files."""
+    try:
+        if file.type == "text/csv":
+            df = pd.read_csv(file)
+            feedback_text = df.iloc[:, 0].dropna().astype(str).tolist()
+            return feedback_text
+        elif file.type == "application/json":
+            json_data = json.load(file)
+            feedback_text = [item.get('feedback', '') for item in json_data if isinstance(item, dict)]
+            return feedback_text
+        elif file.type == "text/plain":
+            return file.getvalue().decode("utf-8").split("\n")
+        else:
+            return ["Unsupported file type"]
+    except Exception as e:
+        st.error(f"Error processing file: {e}")
+        return []
 
-# Display error or feedback extraction status
+# Process uploaded file
 if uploaded_file:
     feedback_text_list = extract_feedback(uploaded_file)
 
-    if feedback_text_list:
+    if feedback_text_list and distilbert_model and bert_topic_model and recommendation_model:
         for feedback_text in feedback_text_list:
-            if st.button(f'Analyze Feedback: "{feedback_text[:30]}..."'):
-                sentiment = distilbert_model.predict([feedback_text])
-                sentiment_result = 'Positive' if sentiment == 1 else 'Negative'
-                st.write(f"Sentiment: {sentiment_result}")
+            with st.expander(f'🔍 Analyze Feedback: "{feedback_text[:30]}..."'):
+                try:
+                    # Sentiment Analysis
+                    sentiment = distilbert_model.predict([feedback_text])
+                    sentiment_result = '😊 Positive' if sentiment == 1 else '☹️ Negative'
+                    st.write(f"**Sentiment:** {sentiment_result}")
+
+                    # Topic Prediction
+                    topics = bert_topic_model.predict([feedback_text])
+                    st.write(f"**Predicted Topic(s):** {topics}")
 
-                topics = bert_topic_model.predict([feedback_text])
-                st.write(f"Predicted Topic(s): {topics}")
+                    # Recommendations
+                    recommendations = recommendation_model.predict([feedback_text])
+                    st.write(f"**Recommended Actions:** {recommendations}")
 
-                recommendations = recommendation_model.predict([feedback_text])
-                st.write(f"Recommended Actions: {recommendations}")
+                except Exception as e:
+                    st.error(f"Error analyzing feedback: {e}")
     else:
-        st.error("Unable to extract feedback from the file.")
+        st.error("⚠️ Unable to analyze feedback. Please check if models are correctly loaded.")
 else:
-    st.info("Please upload a feedback file to analyze.")
+    st.info("📌 Please upload a feedback file to analyze.")