Update app.py

app.py

@@ -9,6 +9,7 @@ import pandas as pd
 import numpy as np
 import joblib
 import tensorflow as tf
+from io import StringIO
 
 # ─── Caching loaders so they only run once per session ───────────────────────
 @st.cache_resource
@@ -34,7 +35,7 @@ def predict_subjects(df_raw):
         if c in df_raw.columns:
             df_raw = df_raw.drop(columns=[c])
 
-    # Re-order to exact feature list
+    # Re‑order to exact feature list
     feature_cols = preprocessor.transformers_[0][2]
     df_features  = df_raw[feature_cols]
 
@@ -48,36 +49,57 @@ def predict_subjects(df_raw):
     df_out = pd.DataFrame({"predicted_subject": labels})
     for i, cls in enumerate(label_encoder.categories_[0]):
         df_out[f"prob_{cls}"] = y_prob[:, i]
+
     return df_out
 
 # ─── Streamlit App ──────────────────────────────────────────────────────────
 def main():
     st.title("🔑 Keystroke Dynamics Authentication")
     st.markdown(
-        "Use the sidebar to enter one row of raw keystroke features, then click **Predict**. "
-        "The model will return the predicted subject ID plus per-class probabilities."
+        """
+        Paste exactly **one** row of comma‑separated feature values (no header) 
+        in the **order below**, then click **Predict**.
+        """
     )
 
-    # Load the feature‐list so we can build inputs
-    preprocessor = load_preprocessor()
-    feature_cols = preprocessor.transformers_[0][2]
+    # Load features list for display and parsing
+    preprocessor  = load_preprocessor()
+    feature_cols  = preprocessor.transformers_[0][2]
 
-    st.sidebar.header("Enter keystroke features")
-    user_vals = {}
-    # one number_input per feature
-    for col in feature_cols:
-        # you can tweak min/max/default as appropriate
-        user_vals[col] = st.sidebar.number_input(col, value=0.0, format="%.4f")
+    st.write("**Feature order:**")
+    st.code(", ".join(feature_cols), language="text")
+
+    # Textarea for single-row input
+    input_text = st.text_area(
+        "Paste your row here (e.g. `0.1491,0.3979,0.2488,...`)", 
+        height=120
+    )
+
+    if st.button("Predict"):
+        if not input_text.strip():
+            st.warning("Please paste one row of comma‑separated values.")
+            return
+
+        # Attempt to parse
+        try:
+            df_input = pd.read_csv(
+                StringIO(input_text.strip()), 
+                header=None, 
+                names=feature_cols
+            )
+        except Exception as e:
+            st.error(f"Could not parse input: {e}")
+            return
 
-    if st.sidebar.button("Predict"):
-        # pack into single-row DataFrame
-        df_input = pd.DataFrame([user_vals], columns=feature_cols)
+        # Ensure exactly one row
+        if df_input.shape[0] != 1:
+            st.error(f"Expected exactly 1 row, but parsed {df_input.shape[0]}.")
+            return
 
-        # Show what we're about to send
-        st.write("### Your input")
+        st.write("### Parsed input")
         st.dataframe(df_input, use_container_width=True)
 
-        # Do prediction
+        # Run prediction
         try:
             df_pred = predict_subjects(df_input)
             st.write("### Prediction")