Update app.py

app.py

@@ -1,30 +1,52 @@
+# app.py - FactoryRAG+: Condition Monitoring with Dashboard, PDF Export, Anomaly Detection & Digital Twin
+
 import streamlit as st
 import pandas as pd
 import numpy as np
+import matplotlib.pyplot as plt
 from sentence_transformers import SentenceTransformer
 from transformers import pipeline
+from sklearn.ensemble import IsolationForest
+import base64
+from io import BytesIO
+from fpdf import FPDF
 
-# Page setup
-st.set_page_config(page_title="FactoryRAG - Upload Logs", layout="wide")
-st.title("🏭 FactoryRAG: Human-Centric AI for Sensor Log Analysis")
+# Streamlit config
+st.set_page_config(page_title="FactoryRAG+ - Smart Sensor Twin", layout="wide")
+st.title("🏭 FactoryRAG+: Smart Dashboard with AI Monitoring, PDF Reporting & Digital Twin")
 
 # Load models
 EMBED_MODEL = SentenceTransformer('sentence-transformers/all-MiniLM-L6-v2')
 GEN_MODEL = pipeline('text2text-generation', model='google/flan-t5-base')
 
-# File uploader
-uploaded_file = st.sidebar.file_uploader("📂 Upload your sensor CSV log file", type=["csv"])
+# Upload data
+uploaded_file = st.sidebar.file_uploader("📂 Upload your condition monitoring CSV", type=["csv"])
 
 if uploaded_file:
     df = pd.read_csv(uploaded_file)
-    st.success("✅ File uploaded and loaded!")
-    st.write("📊 Sensor Data Snapshot:", df.head())
+    numeric_cols = df.select_dtypes(include=np.number).columns.tolist()
+    st.success("✅ Data loaded successfully!")
+
+    st.subheader("📊 Sensor Snapshot")
+    st.dataframe(df.head())
+
+    # --- Multi-signal Dashboard ---
+    st.subheader("📈 Sensor Dashboard")
+    selected_cols = st.multiselect("Select signals to visualize", numeric_cols, default=numeric_cols[:3])
+    fig, ax = plt.subplots(len(selected_cols), 1, figsize=(8, 2 * len(selected_cols)))
+    if len(selected_cols) == 1:
+        ax = [ax]
+    for i, col in enumerate(selected_cols):
+        ax[i].plot(df[col], label=col)
+        ax[i].set_ylabel(col)
+        ax[i].legend()
+    st.pyplot(fig)
 
-    # Convert to chunks
+    # --- Convert Logs to Chunks ---
     def convert_to_chunks(df):
         chunks = []
         for idx, row in df.iterrows():
-            sentence = f"Log entry {idx}: " + ", ".join([f"{col}: {row[col]:.2f}" for col in df.columns])
+            sentence = f"[Log {idx}] " + ", ".join([f"{col}: {row[col]:.2f}" for col in numeric_cols])
             chunks.append(sentence)
         return chunks
 
@@ -32,25 +54,47 @@ if uploaded_file:
         st.session_state.chunks = convert_to_chunks(df)
         st.session_state.embeddings = EMBED_MODEL.encode(st.session_state.chunks)
 
-    # User query
-    query = st.text_input("🔍 Ask something about the sensor logs:")
+    # --- Anomaly Detection ---
+    st.subheader("🚨 Anomaly Detection (Isolation Forest)")
+    iso = IsolationForest(contamination=0.02)
+    anomaly_labels = iso.fit_predict(df[numeric_cols])
+    df['anomaly'] = ['❌' if x == -1 else '' for x in anomaly_labels]
+    st.write("Anomaly Flags:")
+    st.dataframe(df[df['anomaly'] == '❌'].head(5))
 
+    # --- Technical Question Answering ---
+    st.subheader("🧠 Ask Expert Questions")
+    query = st.text_input("Ask a question like 'Where is instability?' or 'Are anomalies visible?'")
     if query:
         query_vec = EMBED_MODEL.encode([query])[0]
-        scores = np.dot(st.session_state.embeddings, query_vec)
-        top_idxs = np.argsort(scores)[-3:][::-1]
+        sims = np.dot(st.session_state.embeddings, query_vec)
+        top_idxs = np.argsort(sims)[-3:][::-1]
         context = "\n".join([st.session_state.chunks[i] for i in top_idxs])
-
-        prompt = f"Answer based on the following logs:\n{context}\n\nQuestion: {query}"
+        prompt = f"Context:\n{context}\n\nAs a reliability engineer, answer: {query}"
         response = GEN_MODEL(prompt, max_length=256)[0]['generated_text']
-
         st.subheader("🤖 FactoryGPT Answer")
-        st.write(response)
+        st.markdown(response)
+
+    # --- Digital Twin Metrics ---
+    st.subheader("🧪 Digital Twin Summary")
+    twin_report = ""
+    for col in selected_cols:
+        max_v = df[col].max()
+        min_v = df[col].min()
+        mean_v = df[col].mean()
+        twin_report += f"{col}\n→ Max: {max_v:.2f}, Min: {min_v:.2f}, Avg: {mean_v:.2f}\n\n"
+    st.code(twin_report)
 
-        st.markdown("### 🧑‍🏭 Human Feedback")
-        st.radio("Is this answer acceptable?", ["Approve", "Correct", "Escalate"], horizontal=True)
+    # --- PDF Export ---
+    st.subheader("📤 Export Digital Twin Report as PDF")
+    pdf = FPDF()
+    pdf.add_page()
+    pdf.set_font("Arial", size=12)
+    pdf.multi_cell(0, 10, f"FactoryRAG+ Digital Twin Report\n\nSelected Signals: {', '.join(selected_cols)}\n\n" + twin_report)
+    pdf_bytes = pdf.output(dest='S').encode('latin1')
+    b64 = base64.b64encode(pdf_bytes).decode()
+    href = f'<a href="data:application/octet-stream;base64,{b64}" download="digital_twin_report.pdf">📄 Download PDF Report</a>'
+    st.markdown(href, unsafe_allow_html=True)
 
-        with st.expander("📄 Retrieved Log Context"):
-            st.code(context)
 else:
-    st.info("👈 Please upload your sensor log file (CSV) to begin.")
+    st.info("👈 Upload a sensor log CSV file to explore digital twin analysis, waveform charts, anomaly detection, and PDF export.")