Add app.py, backend, and model for HF Space

app.py

@@ -1,21 +1,9 @@
 import os
-import subprocess
-
 import joblib
 import pandas as pd
 import streamlit as st
-from backend.train_model import train_model  # your function
-NONE = None
-# from backend.train_model import train_model
+from backend.train_model import train_model
 
-# Get the current directory of the Streamlit script
-# BASE_DIR = os.path.dirname(os.path.abspath(__file__))
-#
-# # Build the absolute path to the model
-# # BASE_DIR = os.path.dirname(os.path.abspath(__file__))
-# MODEL_PATH = os.path.join(BASE_DIR, "..", "models", "best_model.pkl")
-# REPORTS_DIR = os.path.join(BASE_DIR, "..", "reports")
-# PLOTS_DIR = os.path.join(REPORTS_DIR, "plots")
 MODEL_DIR = "models"
 MODEL_FILE = "my_model.pkl"
 MODEL_PATH = os.path.join(MODEL_DIR, MODEL_FILE)
@@ -23,45 +11,35 @@ MODEL_PATH = os.path.join(MODEL_DIR, MODEL_FILE)
 REPORTS_DIR = "reports"
 PLOTS_DIR = os.path.join(REPORTS_DIR, "plots")
 
-
-BACKEND_URL = os.getenv("BACKEND_URL", "http://backend:5000")
+FEATURES = [
+    "Pregnancies", "Glucose", "BloodPressure", "SkinThickness",
+    "Insulin", "BMI", "DiabetesPedigreeFunction", "Age"
+]
 
 st.set_page_config(page_title="Diabetes Prediction Dashboard", layout="wide")
 st.title("🩺 Diabetes Prediction Dashboard")
 
-# ---------- Sidebar ----------
+# Sidebar navigation
 st.sidebar.header("Navigation")
 page = st.sidebar.radio("Go to", ["Predict", "Batch Predict", "Reports & Plots"])
 
-# ---------- Load best model ----------
+model = None
+if os.path.exists(MODEL_PATH):
+    model = joblib.load(MODEL_PATH)
 
-# @st.cache_resource
-# def load_model(path):
-#     if os.path.exists(path):
-#         model = joblib.load(path)
-#         st.sidebar.success("✅ Best model loaded")
-#         return model
-#     else:
-#         result = subprocess.run(["python", "backend/train_model.py"], capture_output=True, text=True)
-#         st.text(result.stdout)
-#         st.text(result.stderr)
-#
-#         # Reload the trained model
-#         model = load_model(MODEL_PATH)
-#         return model
-#
-# model = load_model(MODEL_PATH)
+# ------------------ Train button ------------------
+st.subheader("Train & Predict Diabetes Model")
 
-# ---------- Features ----------
-FEATURES = [
-    "Pregnancies", "Glucose", "BloodPressure", "SkinThickness",
-    "Insulin", "BMI", "DiabetesPedigreeFunction", "Age"
-]
+if st.button("Train Model"):
+    with st.spinner("Training in progress... this may take a while ⏳"):
+        model = train_model()
+        joblib.dump(model, MODEL_PATH)
+    st.success(f"✅ Model trained and saved to `{MODEL_PATH}`")
 
+# ------------------ Predict single ------------------
 def predict_df(df: pd.DataFrame):
-    """Run model prediction on a DataFrame"""
     if model is None:
-        st.error("Model not loaded")
+        st.error("⚠️ Model not loaded. Train first.")
         return None
     missing = [c for c in FEATURES if c not in df.columns]
     if missing:
@@ -69,21 +47,8 @@ def predict_df(df: pd.DataFrame):
         return None
     return model.predict(df[FEATURES])
 
-# # ---------- Pages ----------
-# model = joblib.load(MODEL_PATH)
-st.title("Train & Predict Diabetes Model")
-
-if not os.path.exists(MODEL_PATH):
-    st.warning("No trained model found. Please train the model first.")
-
-if st.button("Train Model"):
-    st.info("Training started...")
-    model = train_model()
-    joblib.dump(model, MODEL_PATH)
-    st.success(f"Model trained and saved to {MODEL_PATH}")
-elif page == "Predict":
+if page == "Predict":
     st.subheader("🔹 Single Prediction")
-
     cols = st.columns(4)
     values = {}
     ranges = {
@@ -92,7 +57,6 @@ elif page == "Predict":
         "Insulin": (0, 900, 80), "BMI": (0.0, 70.0, 25.0),
         "DiabetesPedigreeFunction": (0.0, 3.0, 0.5), "Age": (0, 120, 30)
     }
-
     for i, f in enumerate(FEATURES):
         with cols[i % 4]:
             lo, hi, default = ranges[f]
@@ -100,30 +64,27 @@ elif page == "Predict":
                 values[f] = st.number_input(f, lo, hi, float(default))
             else:
                 values[f] = st.number_input(f, int(lo), int(hi), int(default))
-
     if st.button("Predict"):
         row = pd.DataFrame([values])
         pred = predict_df(row)
         if pred is not None:
             st.success("✅ Diabetic" if int(pred[0]) == 1 else "🟢 Not Diabetic")
 
+# ------------------ Batch predict ------------------
 elif page == "Batch Predict":
     st.subheader("📂 Batch Prediction (Upload CSV)")
     st.caption("CSV must include columns: " + ", ".join(FEATURES))
-
     file = st.file_uploader("Upload CSV", type=["csv"])
     if file is not None:
         df = pd.read_csv(file)
         st.write("Preview of uploaded data:")
         st.dataframe(df.head())
-
         preds = predict_df(df)
         if preds is not None:
             out = df.copy()
             out["Prediction"] = preds
             st.success(f"Predicted {len(out)} rows")
             st.dataframe(out.head())
-
             st.download_button(
                 "⬇️ Download predictions",
                 data=out.to_csv(index=False).encode('utf-8'),
@@ -131,18 +92,9 @@ elif page == "Batch Predict":
                 mime="text/csv"
             )
 
+# ------------------ Reports & plots ------------------
 elif page == "Reports & Plots":
     st.subheader("📊 Model Comparison & Diagnostics")
-
-    # Table report
-    # cmp_path = os.path.join(REPORTS_DIR, "model_comparison.csv")
-    # if os.path.exists(cmp_path):
-    #     cmp_df = pd.read_csv(cmp_path)
-    #     st.dataframe(cmp_df)
-    # else:
-    #     st.warning("⚠️ model_comparison.csv not found. Run training.")
-
-    # Plots grid
     plot_files = [
         ("Accuracy (bar)", "model_accuracy.png"),
         ("F1 (bar)", "model_f1.png"),
@@ -152,7 +104,6 @@ elif page == "Reports & Plots":
         ("LR Loss vs Iterations", "logreg_loss_curves.png"),
         ("LR Accuracy vs Iterations", "logreg_accuracy_curves.png"),
     ]
-
     rows = st.columns(2)
     i = 0
     for title, fname in plot_files:
@@ -160,7 +111,7 @@ elif page == "Reports & Plots":
         if os.path.exists(p):
             with rows[i % 2]:
                 st.markdown(f"**{title}**")
-                st.image(p, use_container_width=True)
+                st.image(p, width=700)  # ✅ works on all Streamlit versions
             i += 1
         else:
             st.info(f"{fname} not available yet.")