Update app.py

app.py

@@ -1,110 +1,129 @@
-import os 
-import nest_asyncio 
-nest_asyncio.apply() 
-import streamlit as st 
-from transformers import pipeline 
-from huggingface_hub import login 
-from streamlit.components.v1 import html 
+import os
+import nest_asyncio
+nest_asyncio.apply()
+import streamlit as st
+from sentence_transformers import CrossEncoder
+from transformers import pipeline
+from huggingface_hub import login
+from streamlit.components.v1 import html
+import pandas as pd
 
 # Retrieve the token from environment variables 
-hf_token = os.environ.get("HF_TOKEN") 
-if not hf_token: 
-    st.error("Hugging Face token not found. Please set the HF_TOKEN environment variable.") 
-    st.stop() 
+hf_token = os.environ.get("HF_TOKEN")
+if not hf_token:
+    st.error("Hugging Face token not found. Please set the HF_TOKEN environment variable.")
+    st.stop()
 
 # Login with the token 
-login(token=hf_token) 
+login(token=hf_token)
 
 # Initialize session state for timer and results 
-if 'result' not in st.session_state: 
-    st.session_state.result = {} 
-if 'timer_started' not in st.session_state: 
-    st.session_state.timer_started = False 
-if 'timer_frozen' not in st.session_state: 
-    st.session_state.timer_frozen = False 
+if 'result' not in st.session_state:
+    st.session_state.result = {}
+if 'timer_started' not in st.session_state:
+    st.session_state.timer_started = False
+if 'timer_frozen' not in st.session_state:
+    st.session_state.timer_frozen = False
 
 # Timer component using HTML and JavaScript 
-def timer(): 
-    return """ 
-    <div id="timer" style="font-size:16px;color:#666;margin-bottom:10px;">⏱️ Elapsed: 00:00</div> 
-    <script> 
-    (function() { 
-    var start = Date.now(); 
-    var timerElement = document.getElementById('timer'); 
-    localStorage.removeItem("freezeTimer"); 
-    var interval = setInterval(function() { 
-    if(localStorage.getItem("freezeTimer") === "true"){ 
-    clearInterval(interval); 
-    timerElement.style.color = '#00cc00'; 
-    return; 
-    } 
-    var elapsed = Date.now() - start; 
-    var minutes = Math.floor(elapsed / 60000); 
-    var seconds = Math.floor((elapsed % 60000) / 1000); 
-    timerElement.innerHTML = '⏱️ Elapsed: ' + 
-    (minutes < 10 ? '0' : '') + minutes + ':' + 
-    (seconds < 10 ? '0' : '') + seconds; 
-    }, 1000); 
-    })(); 
-    </script> 
-    """ 
+def timer():
+    return """
+    <div id="timer" style="font-size:16px;color:#666;margin-bottom:10px;">⏱️ Elapsed: 00:00</div>
+    <script>
+    (function() {
+        var start = Date.now();
+        var timerElement = document.getElementById('timer');
+        localStorage.removeItem("freezeTimer");
+        var interval = setInterval(function() {
+            if(localStorage.getItem("freezeTimer") === "true"){
+                clearInterval(interval);
+                timerElement.style.color = '#00cc00';
+                return;
+            }
+            var elapsed = Date.now() - start;
+            var minutes = Math.floor(elapsed / 60000);
+            var seconds = Math.floor((elapsed % 60000) / 1000);
+            timerElement.innerHTML = '⏱️ Elapsed: ' +
+            (minutes < 10 ? '0' : '') + minutes + ':' +
+            (seconds < 10 ? '0' : '') + seconds;
+        }, 1000);
+    })();
+    </script>
+    """
 
-st.set_page_config(page_title="Sentiment & Report Generator", page_icon="📝") 
-st.header("Sentiment Analysis & Report Generation with Gemma") 
+st.set_page_config(page_title="Sentiment & Report Generator", page_icon="📝")
+st.header("Sentiment Analysis & Report Generation with Gemma")
+
+# Introduction for the Hugging Face interface
+st.write("""
+Welcome to the Sentiment Analysis & Report Generator app!
+This tool leverages Hugging Face's models to analyze the sentiment of your text and generate a detailed report explaining the key insights.
+You can either paste your review text directly into the text area or upload a CSV file containing your reviews.
+""")
 
 # Load models with caching to avoid reloading on every run 
-@st.cache_resource 
-def load_models(): 
-    sentiment_pipe = pipeline("text-classification", model="mixedbread-ai/mxbai-rerank-base-v1") 
-    gemma_pipe = pipeline("text-generation", model="google/gemma-3-1b-it", use_auth_token=hf_token) 
-    return sentiment_pipe, gemma_pipe 
+@st.cache_resource
+def load_models():
+    # Load the sentiment model (CrossEncoder) for ranking sentiment labels.
+    sentiment_model = CrossEncoder("mixedbread-ai/mxbai-rerank-base-v1")
+    # Load the Gemma text generation pipeline.
+    gemma_pipe = pipeline("text-generation", model="google/gemma-3-1b-it", use_auth_token=hf_token)
+    return sentiment_model, gemma_pipe
 
-sentiment_pipe, gemma_pipe = load_models() 
+sentiment_model, gemma_pipe = load_models()
 
-# Provide two options for input: text area or file upload 
-uploaded_file = st.file_uploader("Upload Review File (txt format)", type=["txt"])
+# Provide two options for input: file upload (CSV) or text area
+uploaded_file = st.file_uploader("Upload Review File (CSV format)", type=["csv"])
 user_input = st.text_area("Or, enter your text for sentiment analysis and report generation:")
 
-# If a file is uploaded, override user_input with its contents
 if uploaded_file is not None:
     try:
-        user_input = uploaded_file.read().decode("utf-8")
+        # Read the CSV file; if a column named 'review' exists, use it.
+        df = pd.read_csv(uploaded_file)
+        if 'review' in df.columns:
+            user_input = " ".join(df['review'].astype(str).tolist())
+        else:
+            # Otherwise, join all text from the first column.
+            user_input = " ".join(df.iloc[:, 0].astype(str).tolist())
     except Exception as e:
         st.error(f"Error reading file: {e}")
 
-if st.button("Generate Report"): 
-    if not user_input.strip(): 
+if st.button("Generate Report"):
+    if not user_input.strip():
         st.error("Please enter some text!")
-    else: 
-        if not st.session_state.timer_started and not st.session_state.timer_frozen: 
-            st.session_state.timer_started = True 
-            html(timer(), height=50) 
-        status_text = st.empty() 
-        progress_bar = st.progress(0) 
-        try: 
-            # Stage 1: Sentiment Analysis 
-            status_text.markdown("**🔍 Running sentiment analysis...**") 
-            progress_bar.progress(0) 
-            sentiment_result = sentiment_pipe(user_input) 
-            progress_bar.progress(50) 
-            # Stage 2: Generate Report using Gemma 
-            status_text.markdown("**📝 Generating report with Gemma...**") 
-            prompt = f""" 
-            Generate a detailed report based on the following analysis. 
-            Original text: 
-            "{user_input}" 
-            Sentiment analysis result: 
-            {sentiment_result} 
-            Please provide a concise summary report explaining the sentiment and key insights. 
-            """ 
-            report = gemma_pipe(prompt, max_length=200) 
-            progress_bar.progress(100) 
-            status_text.success("**✅ Generation complete!**") 
-            html("<script>localStorage.setItem('freezeTimer', 'true');</script>", height=0) 
-            st.session_state.timer_frozen = True 
-            st.write("**Sentiment Analysis Result:**", sentiment_result) 
-            st.write("**Generated Report:**", report[0]['generated_text']) 
-        except Exception as e: 
-            html("<script>document.getElementById('timer').remove();</script>") 
-            status_text.error(f"**❌ Error:** {str(e)}") 
+    else:
+        if not st.session_state.timer_started and not st.session_state.timer_frozen:
+            st.session_state.timer_started = True
+            html(timer(), height=50)
+        status_text = st.empty()
+        progress_bar = st.progress(0)
+        try:
+            # Stage 1: Sentiment Analysis using CrossEncoder ranking
+            status_text.markdown("**🔍 Running sentiment analysis...**")
+            progress_bar.progress(0)
+            # Use sentiment analysis as ranking over sentiment labels.
+            labels = ["positive", "neutral", "negative"]
+            sentiment_result = sentiment_model.rank(user_input, labels, return_documents=True, top_k=1)
+            progress_bar.progress(50)
+            
+            # Stage 2: Generate Report using Gemma
+            status_text.markdown("**📝 Generating report with Gemma...**")
+            prompt = f"""
+Generate a detailed report based on the following analysis.
+Original text:
+"{user_input}"
+Sentiment analysis result:
+{sentiment_result}
+Please provide a concise summary report explaining the sentiment and key insights.
+"""
+            report = gemma_pipe(prompt, max_length=200)
+            progress_bar.progress(100)
+            status_text.success("**✅ Generation complete!**")
+            html("<script>localStorage.setItem('freezeTimer', 'true');</script>", height=0)
+            st.session_state.timer_frozen = True
+            st.write("**Sentiment Analysis Result:**", sentiment_result)
+            st.write("**Generated Report:**", report[0]['generated_text'])
+        except Exception as e:
+            html("<script>document.getElementById('timer').remove();</script>")
+            status_text.error(f"**❌ Error:** {str(e)}")
             progress_bar.empty()