Update app.py

app.py

@@ -10,36 +10,23 @@ import PyPDF2
 import tempfile
 import os
 
-st.set_page_config(page_title="🚀 WiFi Anomaly Detection", layout="wide")
-
 # -------------------------------
-# WiFi Anomaly Detection Overview
+# Page Configuration and Header
 # -------------------------------
-st.title("🚀 WiFi Anomaly Detection Overview")
+st.set_page_config(page_title="🚀 WiFi Anomaly Detection", layout="wide")
+st.title("🚀 WiFi Anomaly Detection System")
 st.markdown("""
-**Detect anomalies in Public Wi-Fi Systems**:  
-Identify suspicious spikes that may indicate hacking attempts, ensuring proactive maintenance and reliable network performance.
+> "Innovation distinguishes between a leader and a follower." – *Steve Jobs*
+
+> "The future depends on what you do today." – *Mahatma Gandhi*
 """)
-st.markdown("### How it Works:")
 st.markdown("""
-- **Data Collection:** Upload network logs in CSV, TXT, or PDF format.
-- **Anomaly Detection:** Use AI algorithms to automatically spot unusual patterns.
-- **Visualization:** Review data in 2D and 3D interactive charts.
-- **Report Generation:** Download a comprehensive PDF report with summaries and visuals.
+Welcome to the WiFi Anomaly Detection System. This application uses AI to proactively detect abnormal behavior in Public Wi-Fi systems, identifying suspicious spikes that may indicate hacking attempts. Let’s build a more secure network, one anomaly at a time!
 """)
 
 # -------------------------------
-# Sidebar: File Upload & Options
-# -------------------------------
-st.sidebar.header("📁 Upload Data File")
-uploaded_file = st.sidebar.file_uploader("Choose a file", type=["csv", "txt", "pdf"])
-st.sidebar.markdown("---")
-model_option = st.sidebar.radio("Select Anomaly Detection Model", ("Local Model", "Groq API"))
-
+# Define Helper Functions
 # -------------------------------
-# Helper Functions
-# -------------------------------
-
 def load_data(uploaded_file):
     file_type = uploaded_file.name.split('.')[-1].lower()
     if file_type == 'csv':
@@ -51,7 +38,6 @@ def load_data(uploaded_file):
             return None, None
     elif file_type == 'txt':
         try:
-            # Try comma separated first; if not, try whitespace separation
             try:
                 df = pd.read_csv(uploaded_file, sep=",")
             except:
@@ -66,7 +52,6 @@ def load_data(uploaded_file):
             text = ""
             for page in pdf_reader.pages:
                 text += page.extract_text()
-            # For demonstration, create a DataFrame with one text column
             df = pd.DataFrame({"text": [text]})
             return df, "pdf"
         except Exception as e:
@@ -85,23 +70,18 @@ def run_local_anomaly_detection(df):
     X = df[numeric_cols].fillna(0)
     model = IsolationForest(contamination=0.1, random_state=42)
     model.fit(X)
-    # Model returns -1 for anomalies, 1 for normal records
     df['anomaly'] = model.predict(X)
     df['anomaly_flag'] = df['anomaly'].apply(lambda x: "🚨 Anomaly" if x == -1 else "✅ Normal")
     return df
 
 def call_groq_api(df):
     # ----- Dummy Groq API integration -----
-    # In a real implementation, you would send your data via a POST request like:
-    #   response = requests.post("https://api.groq.ai/detect", json=df.to_dict(orient="records"))
-    # and then process the JSON response.
-    # For demo purposes, we simply call the local model.
-    # ----------------------------------------
+    # Replace this dummy call with an actual Groq API call as needed.
     df = run_local_anomaly_detection(df)
     return df
 
 def generate_plots(df):
-    # Create 2D and 3D scatter plots based on the first numeric columns
+    # Generate 2D and 3D plots from the first numeric columns
     numeric_cols = df.select_dtypes(include=[np.number]).columns
     fig2d, fig3d = None, None
     if len(numeric_cols) >= 2:
@@ -144,53 +124,99 @@ def generate_pdf_report(summary_text, fig2d, fig3d):
     for image in image_files:
         os.remove(image)
     
-    pdf_output = io.BytesIO()
-    pdf.output(pdf_output)
-    pdf_data = pdf_output.getvalue()
-    pdf_output.close()
+    pdf_data = pdf.output(dest="S").encode("latin1")
     return pdf_data
 
 # -------------------------------
-# Main Workflow
+# Initialize Session State Variables
 # -------------------------------
-if uploaded_file is not None:
-    df, file_type = load_data(uploaded_file)
-    
-    if df is not None:
-        if file_type == "pdf":
-            st.subheader("📄 Extracted Text from PDF:")
-            st.text_area("PDF Content", df["text"][0], height=300)
-        else:
-            st.subheader("📊 Data Preview:")
-            st.dataframe(df.head())
-        
-        if st.button("▶️ Check Data Visualization & Summary"):
-            if file_type in ["csv", "txt"]:
-                # Run the selected anomaly detection method
-                if model_option == "Local Model":
-                    df_result = run_local_anomaly_detection(df)
-                else:
-                    df_result = call_groq_api(df)
-                    
-                st.subheader("📝 Anomaly Detection Summary:")
-                anomaly_count = (df_result['anomaly'] == -1).sum()
-                total_count = df_result.shape[0]
-                summary_text = f"Total records: {total_count}\nDetected anomalies: {anomaly_count}"
-                st.text(summary_text)
-                st.dataframe(df_result.head())
-                
-                fig2d, fig3d = generate_plots(df_result)
-                if fig2d:
-                    st.plotly_chart(fig2d, use_container_width=True)
-                if fig3d:
-                    st.plotly_chart(fig3d, use_container_width=True)
-                
-                if st.button("⬇️ Download Report as PDF"):
-                    pdf_data = generate_pdf_report(summary_text, fig2d, fig3d)
-                    st.download_button("Download PDF", data=pdf_data,
-                                       file_name="wifi_anomaly_report.pdf",
-                                       mime="application/pdf")
+if "step" not in st.session_state:
+    st.session_state.step = "upload"
+if "df" not in st.session_state:
+    st.session_state.df = None
+if "df_processed" not in st.session_state:
+    st.session_state.df_processed = None
+if "fig2d" not in st.session_state:
+    st.session_state.fig2d = None
+if "fig3d" not in st.session_state:
+    st.session_state.fig3d = None
+if "summary_text" not in st.session_state:
+    st.session_state.summary_text = ""
+
+# -------------------------------
+# Sidebar: Step Buttons
+# -------------------------------
+st.sidebar.title("🔧 Application Steps")
+if st.sidebar.button("📁 Upload File"):
+    st.session_state.step = "upload"
+if st.sidebar.button("📊 Data Visualization"):
+    st.session_state.step = "viz"
+if st.sidebar.button("📈 Statistic Analysis"):
+    st.session_state.step = "stats"
+if st.sidebar.button("⬇️ Download Report"):
+    st.session_state.step = "download"
+
+# -------------------------------
+# Main Workflow Based on Step
+# -------------------------------
+if st.session_state.step == "upload":
+    st.subheader("Step 1: Upload Your Data File")
+    st.markdown("Please upload a CSV, TXT, or PDF file with network data. The expected columns for CSV/TXT files are:")
+    st.code("['traffic', 'latency', 'packet_loss']", language="python")
+    uploaded_file = st.file_uploader("Choose a file", type=["csv", "txt", "pdf"])
+    if uploaded_file is not None:
+        df, file_type = load_data(uploaded_file)
+        if df is not None:
+            st.session_state.df = df
+            st.success("File uploaded and processed successfully!")
+            if file_type == "pdf":
+                st.subheader("Extracted Text from PDF:")
+                st.text_area("PDF Content", df["text"][0], height=300)
             else:
-                st.info("Anomaly detection is available only for CSV/TXT data.")
-else:
-    st.info("Please upload a CSV, TXT, or PDF file to begin. 😊")
+                st.subheader("Data Preview:")
+                st.dataframe(df.head())
+    else:
+        st.info("Awaiting file upload. 😊")
+
+elif st.session_state.step == "viz":
+    st.subheader("Step 2: Data Visualization")
+    if st.session_state.df is None:
+        st.error("Please upload a file first in the 'Upload File' step.")
+    else:
+        # Process the data if not already done
+        if st.session_state.df_processed is None:
+            # Here, you can choose between the local model or Groq API; we use the local model for demo.
+            st.session_state.df_processed = run_local_anomaly_detection(st.session_state.df)
+        fig2d, fig3d = generate_plots(st.session_state.df_processed)
+        st.session_state.fig2d = fig2d
+        st.session_state.fig3d = fig3d
+        if fig2d:
+            st.plotly_chart(fig2d, use_container_width=True)
+        if fig3d:
+            st.plotly_chart(fig3d, use_container_width=True)
+
+elif st.session_state.step == "stats":
+    st.subheader("Step 3: Statistic Analysis")
+    if st.session_state.df_processed is None:
+        st.error("Data has not been processed yet. Please complete the Data Visualization step first.")
+    else:
+        df_result = st.session_state.df_processed
+        anomaly_count = (df_result['anomaly'] == -1).sum()
+        total_count = df_result.shape[0]
+        st.session_state.summary_text = f"Total records: {total_count}\nDetected anomalies: {anomaly_count}"
+        st.markdown("**Anomaly Detection Summary:**")
+        st.text(st.session_state.summary_text)
+        st.markdown("**Detailed Data:**")
+        st.dataframe(df_result.head())
+        st.markdown("**Descriptive Statistics:**")
+        st.dataframe(df_result.describe())
+
+elif st.session_state.step == "download":
+    st.subheader("Step 4: Download PDF Report")
+    if st.session_state.df_processed is None or (st.session_state.fig2d is None and st.session_state.fig3d is None):
+        st.error("Please complete the previous steps (Upload, Visualization, Statistic Analysis) before downloading the report.")
+    else:
+        pdf_data = generate_pdf_report(st.session_state.summary_text, st.session_state.fig2d, st.session_state.fig3d)
+        st.download_button("⬇️ Download PDF Report", data=pdf_data,
+                           file_name="wifi_anomaly_report.pdf",
+                           mime="application/pdf")