Update app.py

app.py

@@ -12,7 +12,6 @@ RATING_MAP = {
     2: "Positive (⭐⭐⭐)"
 }
 
-# Load models
 @st.cache_resource
 def load_models():
     sentiment_model = pipeline(
@@ -22,23 +21,20 @@ def load_models():
     ner_model = pipeline("ner", model="dslim/bert-base-NER")
     return sentiment_model, ner_model
 
-# Sentiment analysis
 def analyze_sentiment(text, model):
     result = model(text)[0]
-    rating = int(result['label'].split('_')[-1])  # Get 0, 1, or 2
+    rating = int(result['label'].split('_')[-1])
     return {
         'rating': rating,
         'label': RATING_MAP[rating],
         'score': result['score']
     }
 
-# Entity extraction
 def extract_aspects(text, model):
     entities = model(text)
     aspects = []
     current_entity = ""
     
-    # Merge subword tokens
     for entity in entities:
         if entity['word'].startswith('##'):
             current_entity += entity['word'][2:]
@@ -59,7 +55,6 @@ def extract_aspects(text, model):
     
     return [a for a in aspects if a['type'] in ['PRODUCT', 'ORG', 'PERSON']]
 
-# Visualization functions
 def plot_sentiment_distribution(df):
     fig, ax = plt.subplots()
     df['label'].value_counts().loc[list(RATING_MAP.values())].plot.pie(
@@ -84,41 +79,32 @@ def plot_wordcloud(negative_reviews):
     ax.axis('off')
     return fig
 
-# Main app
 def main():
     st.title("Restaurant Review Analyzer")
     st.markdown("Using fine-tuned model for sentiment and aspect analysis")
     
-    # Initialize models
     sentiment_model, ner_model = load_models()
     
-    # Sidebar controls
     st.sidebar.header("Analysis Options")
     analysis_mode = st.sidebar.radio(
         "Select Mode",
         ["Single Review", "Batch Analysis"]
     )
     
-    # Initialize session state
     if 'history' not in st.session_state:
         st.session_state.history = pd.DataFrame(
             columns=['text', 'rating', 'label', 'date', 'aspects']
         )
     
     if analysis_mode == "Single Review":
-        # Single review analysis
         user_input = st.text_area("Enter or paste a restaurant review:", height=150)
         
         if st.button("Analyze"):
             if user_input:
                 with st.spinner("Analyzing..."):
-                    # Sentiment analysis
                     sentiment = analyze_sentiment(user_input, sentiment_model)
-                    
-                    # Aspect extraction
                     aspects = extract_aspects(user_input, ner_model)
                     
-                    # Save to history
                     new_entry = pd.DataFrame([{
                         'text': user_input,
                         'rating': sentiment['rating'],
@@ -131,7 +117,6 @@ def main():
                         ignore_index=True
                     )
                     
-                    # Display results
                     st.subheader("Analysis Results")
                     col1, col2 = st.columns(2)
                     with col1:
@@ -149,7 +134,6 @@ def main():
                 st.warning("Please enter a review")
     
     else:
-        # Batch analysis mode
         uploaded_file = st.file_uploader("Upload CSV file", type=["csv"])
         
         if uploaded_file:
@@ -181,21 +165,17 @@ def main():
                     )
                     st.success(f"Completed analysis of {len(df)} reviews")
     
-    # Display historical data and visualizations
     if not st.session_state.history.empty:
         st.divider()
         st.header("Analysis History")
         
-        # Raw data display
         with st.expander("View Raw Data"):
             st.dataframe(st.session_state.history)
         
-        # Visualizations
         st.subheader("Sentiment Distribution")
         fig1 = plot_sentiment_distribution(st.session_state.history)
         st.pyplot(fig1)
         
-        # Negative reviews word cloud
         negative_reviews = st.session_state.history[
             st.session_state.history['rating'] == 0
         ]['text'].tolist()
@@ -207,7 +187,6 @@ def main():
         else:
             st.info("No negative reviews yet")
         
-        # Time trend analysis
         if len(st.session_state.history) > 1:
             st.subheader("Rating Trend Over Time")
             time_df = st.session_state.history.copy()