Update app.py

app.py

@@ -1,6 +1,6 @@
 # app.py — ST_Min_Horizontal_Stress (σhmin)
-# End-to-end Streamlit app that loads MODEL/META from Hugging Face Hub using hf_hub_download
-# (handles large LFS files; uses AUTH_TOKEN if repo is private). Falls back to local/mnt/data.
+# Streamlit app that LOADS THE MODEL/META FROM USER UPLOADS (memory only; no auth, no saving).
+# After the model is in memory, the rest of the workflow (Train/Test/Validate/Predict) is unchanged.
 
 import io, json, os, base64, math
 from pathlib import Path
@@ -16,12 +16,10 @@ import matplotlib
 matplotlib.use("Agg")
 import matplotlib.pyplot as plt
 from matplotlib.ticker import FuncFormatter
-from huggingface_hub import hf_hub_download
 
 import plotly.graph_objects as go
 from sklearn.metrics import mean_squared_error
 
-
 # =========================
 # App constants / defaults
 # =========================
@@ -44,12 +42,6 @@ BOLD_FONT = "Arial Black, Arial, sans-serif"
 
 STRICT_VERSION_CHECK = True
 
-# Local fallbacks (optional) — used only if HF Hub fails, or if you upload to /mnt/data
-MODELS_DIR = Path("models")
-DEFAULT_MODEL = MODELS_DIR / "minstress_model.joblib"
-MODEL_FALLBACKS = [MODELS_DIR / "model.joblib", MODELS_DIR / "model.pkl"]
-META_CANDIDATES = [MODELS_DIR / "minstress_meta.json", MODELS_DIR / "meta.json"]
-
 # =========================
 # Page / CSS
 # =========================
@@ -79,7 +71,7 @@ TABLE_CENTER_CSS = [
 ]
 
 # =========================
-# Password gate
+# Password gate (optional)
 # =========================
 def inline_logo(path="logo.png") -> str:
     try:
@@ -90,15 +82,14 @@ def inline_logo(path="logo.png") -> str:
         return ""
 
 def add_password_gate() -> None:
-    # Read password from Secrets or ENV
     try:
         required = st.secrets.get("APP_PASSWORD", "")
     except Exception:
         required = os.environ.get("APP_PASSWORD", "")
 
     if not required:
-        st.warning("Set APP_PASSWORD in Secrets (or environment) and restart.")
-        st.stop()
+        return  # disable gate if no password set
+
     if st.session_state.get("auth_ok", False):
         return
 
@@ -214,347 +205,6 @@ def _make_X(df: pd.DataFrame, features: list[str]) -> pd.DataFrame:
         X[c] = pd.to_numeric(X[c], errors="coerce")
     return X
 
-# =========================
-# Export helpers
-# =========================
-def _summary_table(df: pd.DataFrame, cols: list[str]) -> pd.DataFrame:
-    cols = [c for c in cols if c in df.columns]
-    if not cols: return pd.DataFrame()
-    tbl = (df[cols]
-           .agg(['min','max','mean','std'])
-           .T.rename(columns={"min":"Min","max":"Max","mean":"Mean","std":"Std"})
-           .reset_index(names="Field"))
-    return _round_numeric(tbl, 3)
-
-def _train_ranges_df(ranges: dict[str, tuple[float, float]]) -> pd.DataFrame:
-    if not ranges: return pd.DataFrame()
-    df = pd.DataFrame(ranges).T.reset_index()
-    df.columns = ["Feature", "Min", "Max"]
-    return _round_numeric(df, 3)
-
-def _excel_autofit(writer, sheet_name: str, df: pd.DataFrame, min_w: int = 8, max_w: int = 40):
-    try:
-        import xlsxwriter  # noqa: F401
-    except Exception:
-        return
-    ws = writer.sheets[sheet_name]
-    for i, col in enumerate(df.columns):
-        series = df[col].astype(str)
-        max_len = max([len(str(col))] + series.map(len).tolist())
-        ws.set_column(i, i, max(min_w, min(max_len + 2, max_w)))
-    ws.freeze_panes(1, 0)
-
-def _available_sections() -> list[str]:
-    res = st.session_state.get("results", {})
-    sections = []
-    if "Train" in res:       sections += ["Training","Training_Metrics","Training_Summary"]
-    if "Test" in res:        sections += ["Testing","Testing_Metrics","Testing_Summary"]
-    if "Validate" in res:    sections += ["Validation","Validation_Metrics","Validation_Summary","Validation_OOR"]
-    if "PredictOnly" in res: sections += ["Prediction","Prediction_Summary"]
-    if st.session_state.get("train_ranges"): sections += ["Training_Ranges"]
-    sections += ["Info"]
-    return sections
-
-def build_export_workbook(selected: list[str], ndigits: int = 3, do_autofit: bool = True) -> tuple[bytes|None, str|None, list[str]]:
-    res = st.session_state.get("results", {})
-    if not res: return None, None, []
-    sheets: dict[str, pd.DataFrame] = {}
-    order: list[str] = []
-
-    def _add(name: str, df: pd.DataFrame):
-        if df is None or (isinstance(df, pd.DataFrame) and df.empty): return
-        sheets[name] = _round_numeric(df, ndigits); order.append(name)
-
-    if "Training" in selected and "Train" in res: _add("Training", res["Train"])
-    if "Training_Metrics" in selected and res.get("m_train"): _add("Training_Metrics", pd.DataFrame([res["m_train"]]))
-    if "Training_Summary" in selected and "Train" in res:
-        tr_cols = FEATURES + [c for c in [TARGET, PRED_COL] if c in res["Train"].columns]
-        _add("Training_Summary", _summary_table(res["Train"], tr_cols))
-
-    if "Testing" in selected and "Test" in res: _add("Testing", res["Test"])
-    if "Testing_Metrics" in selected and res.get("m_test"): _add("Testing_Metrics", pd.DataFrame([res["m_test"]]))
-    if "Testing_Summary" in selected and "Test" in res:
-        te_cols = FEATURES + [c for c in [TARGET, PRED_COL] if c in res["Test"].columns]
-        _add("Testing_Summary", _summary_table(res["Test"], te_cols))
-
-    if "Validation" in selected and "Validate" in res: _add("Validation", res["Validate"])
-    if "Validation_Metrics" in selected and res.get("m_val"): _add("Validation_Metrics", pd.DataFrame([res["m_val"]]))
-    if "Validation_Summary" in selected and res.get("sv_val"): _add("Validation_Summary", pd.DataFrame([res["sv_val"]]))
-    if "Validation_OOR" in selected and isinstance(res.get("oor_tbl"), pd.DataFrame) and not res["oor_tbl"].empty:
-        _add("Validation_OOR", res["oor_tbl"].reset_index(drop=True))
-
-    if "Prediction" in selected and "PredictOnly" in res: _add("Prediction", res["PredictOnly"])
-    if "Prediction_Summary" in selected and res.get("sv_pred"): _add("Prediction_Summary", pd.DataFrame([res["sv_pred"]]))
-
-    if "Training_Ranges" in selected and st.session_state.get("train_ranges"):
-        _add("Training_Ranges", _train_ranges_df(st.session_state["train_ranges"]))
-
-    if "Info" in selected:
-        info = pd.DataFrame([
-            {"Key": "AppName",    "Value": APP_NAME},
-            {"Key": "Tagline",    "Value": TAGLINE},
-            {"Key": "Target",     "Value": TARGET},
-            {"Key": "PredColumn", "Value": PRED_COL},
-            {"Key": "Features",   "Value": ", ".join(FEATURES)},
-            {"Key": "ExportedAt", "Value": datetime.now().strftime("%Y-%m-%d %H:%M:%S")},
-        ])
-        _add("Info", info)
-
-    if not order: return None, None, []
-
-    bio = io.BytesIO()
-    engine = _excel_engine()
-    with pd.ExcelWriter(bio, engine=engine) as writer:
-        for name in order:
-            df = sheets[name]; sheet = _excel_safe_name(name)
-            df.to_excel(writer, sheet_name=sheet, index=False)
-            if do_autofit: _excel_autofit(writer, sheet, df)
-    bio.seek(0)
-    fname = f"MinStress_Export_{datetime.now().strftime('%Y%m%d_%H%M%S')}.xlsx"
-    return bio.getvalue(), fname, order
-
-def render_export_button(phase_key: str) -> None:
-    res = st.session_state.get("results", {})
-    if not res: return
-    st.divider()
-    st.markdown("### Export to Excel")
-    options = _available_sections()
-    selected_sheets = st.multiselect(
-        "Sheets to include",
-        options=options,
-        default=[],
-        placeholder="Choose option(s)",
-        help="Pick the sheets you want in the Excel export.",
-        key=f"sheets_{phase_key}",
-    )
-    if not selected_sheets:
-        st.caption("Select one or more sheets above to enable export.")
-        st.download_button("⬇️ Export Excel", data=b"", file_name="MinStress_Export.xlsx",
-                           mime="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
-                           disabled=True, key=f"download_{phase_key}")
-        return
-    data, fname, names = build_export_workbook(selected=selected_sheets, ndigits=3, do_autofit=True)
-    if names: st.caption("Will include: " + ", ".join(names))
-    st.download_button("⬇️ Export Excel", data=(data or b""), file_name=(fname or "MinStress_Export.xlsx"),
-                       mime="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
-                       disabled=(data is None), key=f"download_{phase_key}")
-
-# =========================
-# Plots
-# =========================
-def cross_plot_static(actual, pred):
-    a = pd.Series(actual, dtype=float)
-    p = pd.Series(pred,   dtype=float)
-    lo = float(min(a.min(), p.min())); hi = float(max(a.max(), p.max()))
-    pad = 0.03 * (hi - lo if hi > lo else 1.0)
-    lo2, hi2 = lo - pad, hi + pad
-    ticks = np.linspace(lo2, hi2, 5)
-
-    dpi = 110
-    fig, ax = plt.subplots(figsize=(CROSS_W / dpi, CROSS_H / dpi), dpi=dpi, constrained_layout=False)
-    ax.scatter(a, p, s=14, c=COLORS["pred"], alpha=0.9, linewidths=0)
-    ax.plot([lo2, hi2], [lo2, hi2], linestyle="--", linewidth=1.2, color=COLORS["ref"])
-
-    ax.set_xlim(lo2, hi2); ax.set_ylim(lo2, hi2)
-    ax.set_xticks(ticks);  ax.set_yticks(ticks)
-    ax.set_aspect("equal", adjustable="box")
-
-    fmt = FuncFormatter(lambda x, _: f"{x:.2f}")
-    ax.xaxis.set_major_formatter(fmt); ax.yaxis.set_major_formatter(fmt)
-
-    ax.set_xlabel(f"Actual Min Stress ({UNITS})",  fontweight="bold", fontsize=10, color="black")
-    ax.set_ylabel(f"Predicted Min Stress ({UNITS})", fontweight="bold", fontsize=10, color="black")
-    ax.tick_params(labelsize=6, colors="black")
-    ax.grid(True, linestyle=":", alpha=0.3)
-    for spine in ax.spines.values():
-        spine.set_linewidth(1.1); spine.set_color("#444")
-
-    fig.subplots_adjust(left=0.16, bottom=0.16, right=0.98, top=0.98)
-    return fig
-
-def track_plot(df, include_actual=True):
-    depth_col = next((c for c in df.columns if 'depth' in str(c).lower()), None)
-    if depth_col is not None:
-        y = pd.to_numeric(df[depth_col], errors="coerce"); ylab = depth_col
-        y_range = [float(np.nanmax(y)), float(np.nanmin(y))]  # reversed
-    else:
-        y = pd.Series(np.arange(1, len(df) + 1)); ylab = "Point Index"
-        y_range = [float(y.max()), float(y.min())]
-
-    x_series = pd.Series(df.get(PRED_COL, pd.Series(dtype=float))).astype(float)
-    act_col = ACTUAL_COL if (ACTUAL_COL and ACTUAL_COL in df.columns) else TARGET
-    if include_actual and act_col in df.columns:
-        x_series = pd.concat([x_series, pd.Series(df[act_col]).astype(float)], ignore_index=True)
-    x_lo, x_hi = float(x_series.min()), float(x_series.max())
-    x_pad = 0.03 * (x_hi - x_lo if x_hi > x_lo else 1.0)
-    xmin, xmax = x_lo - x_pad, x_hi + x_pad
-    tick0 = _nice_tick0(xmin, step=max((xmax - xmin) / 10.0, 0.1))
-
-    fig = go.Figure()
-    if PRED_COL in df.columns:
-        fig.add_trace(go.Scatter(
-            x=df[PRED_COL], y=y, mode="lines",
-            line=dict(color=COLORS["pred"], width=1.8),
-            name=PRED_COL,
-            hovertemplate=f"{PRED_COL}: "+"%{x:.2f}<br>"+ylab+": %{y}<extra></extra>"
-        ))
-    if include_actual and act_col in df.columns:
-        fig.add_trace(go.Scatter(
-            x=df[act_col], y=y, mode="lines",
-            line=dict(color=COLORS["actual"], width=2.0, dash="dot"),
-            name=f"{act_col} (actual)",
-            hovertemplate=f"{act_col}: "+"%{x:.2f}<br>"+ylab+": %{y}<extra></extra>"
-        ))
-
-    fig.update_layout(
-        height=TRACK_H, width=TRACK_W, autosize=False,
-        paper_bgcolor="#fff", plot_bgcolor="#fff",
-        margin=dict(l=64, r=16, t=36, b=48), hovermode="closest",
-        font=dict(size=FONT_SZ, color="#000"),
-        legend=dict(x=0.98, y=0.05, xanchor="right", yanchor="bottom",
-                    bgcolor="rgba(255,255,255,0.75)", bordercolor="#ccc", borderwidth=1),
-        legend_title_text=""
-    )
-    fig.update_xaxes(
-        title_text=f"Min Stress ({UNITS})",
-        title_font=dict(size=20, family=BOLD_FONT, color="#000"),
-        tickfont=dict(size=15, family=BOLD_FONT, color="#000"),
-        side="top", range=[xmin, xmax],
-        ticks="outside", tickformat=",.2f", tickmode="auto", tick0=tick0,
-        showline=True, linewidth=1.2, linecolor="#444", mirror=True,
-        showgrid=True, gridcolor="rgba(0,0,0,0.12)", automargin=True
-    )
-    fig.update_yaxes(
-        title_text=ylab,
-        title_font=dict(size=20, family=BOLD_FONT, color="#000"),
-        tickfont=dict(size=15, family=BOLD_FONT, color="#000"),
-        range=y_range, ticks="outside",
-        showline=True, linewidth=1.2, linecolor="#444", mirror=True,
-        showgrid=True, gridcolor="rgba(0,0,0,0.12)", automargin=True
-    )
-    return fig
-
-def preview_tracks(df: pd.DataFrame, cols: list[str]):
-    cols = [c for c in cols if c in df.columns]
-    n = len(cols)
-    if n == 0:
-        fig, ax = plt.subplots(figsize=(4, 2))
-        ax.text(0.5, 0.5, "No selected columns", ha="center", va="center")
-        ax.axis("off")
-        return fig
-
-    depth_col = next((c for c in df.columns if 'depth' in str(c).lower()), None)
-    if depth_col is not None:
-        idx = pd.to_numeric(df[depth_col], errors="coerce")
-        y_label = depth_col
-        y_min, y_max = float(np.nanmin(idx)), float(np.nanmax(idx))
-    else:
-        idx = pd.Series(np.arange(1, len(df) + 1))
-        y_label = "Point Index"
-        y_min, y_max = float(idx.min()), float(idx.max())
-
-    cmap = plt.get_cmap("tab20")
-    col_colors = {col: cmap(i % cmap.N) for i, col in enumerate(cols)}
-
-    fig, axes = plt.subplots(1, n, figsize=(2.4 * n, 7.0), sharey=True, dpi=100)
-    if n == 1:
-        axes = [axes]
-
-    for i, (ax, col) in enumerate(zip(axes, cols)):
-        x = pd.to_numeric(df[col], errors="coerce")
-        ax.plot(x, idx, '-', lw=1.6, color=col_colors[col])
-        ax.set_xlabel(col); ax.xaxis.set_label_position('top'); ax.xaxis.tick_top()
-        ax.set_ylim(y_max, y_min)  # reversed depth down
-        ax.grid(True, linestyle=":", alpha=0.3)
-        if i == 0:
-            ax.set_ylabel(y_label)
-        else:
-            ax.tick_params(labelleft=False); ax.set_ylabel("")
-
-    fig.tight_layout()
-    return fig
-
-# =========================
-# HF Hub loaders (primary) + local fallbacks
-# =========================
-def _hf_repo_id() -> str:
-    # Defaults to your repo; can be overridden via secrets/env
-    return (
-        os.environ.get("HF_REPO_ID", "")
-        or (st.secrets.get("HF_REPO_ID", "") if hasattr(st, "secrets") else "")
-        or "Smart-Thinking/minstress_model"
-    )
-
-def _hf_token() -> str:
-    return os.environ.get("AUTH_TOKEN", "") or (st.secrets.get("AUTH_TOKEN", "") if hasattr(st, "secrets") else "")
-
-def _model_file() -> str:
-    return (os.environ.get("MODEL_FILE", "")
-            or (st.secrets.get("MODEL_FILE", "") if hasattr(st, "secrets") else "")
-            or "minstress_model/minstress_model.joblib")
-
-def _meta_file() -> str:
-    return (os.environ.get("META_FILE", "")
-            or (st.secrets.get("META_FILE", "") if hasattr(st, "secrets") else "")
-            or "minstress_model/minstress_meta.json")
-
-@st.cache_resource(show_spinner=False)
-def load_model_from_hub_or_local() -> object:
-    repo_id = _hf_repo_id()
-    filename = _model_file()
-    token = _hf_token()
-
-    # Try HF Hub (works with LFS/large files)
-    try:
-        model_path = hf_hub_download(repo_id=repo_id, filename=filename, token=token)
-        return joblib.load(model_path)
-    except Exception as e:
-        st.warning(f"HF Hub model download failed ({repo_id}/{filename}): {e}. Trying local fallbacks…")
-
-    # Local fallbacks in repo
-    for p in [DEFAULT_MODEL, *MODEL_FALLBACKS]:
-        if p.exists() and p.stat().st_size > 0:
-            return joblib.load(str(p))
-
-    # /mnt/data fallback (manual upload for quick testing)
-    p2 = Path("/mnt/data/minstress_model.joblib")
-    if p2.exists() and p2.stat().st_size > 0:
-        return joblib.load(str(p2))
-
-    raise FileNotFoundError("No model available via HF Hub or local fallbacks.")
-
-@st.cache_data(show_spinner=False)
-def load_meta_from_hub_or_local() -> dict:
-    repo_id = _hf_repo_id()
-    filename = _meta_file()
-    token = _hf_token()
-
-    # Try HF Hub
-    try:
-        meta_path = hf_hub_download(repo_id=repo_id, filename=filename, token=token)
-        with open(meta_path, "r", encoding="utf-8") as f:
-            return json.load(f)
-    except Exception as e:
-        st.warning(f"HF Hub meta download failed ({repo_id}/{filename}): {e}. Trying local fallbacks…")
-
-    # Local fallbacks (repo)
-    for p in META_CANDIDATES:
-        if p.exists():
-            try:
-                return json.loads(p.read_text(encoding="utf-8"))
-            except Exception:
-                pass
-
-    # /mnt/data fallback
-    p2 = Path("/mnt/data/minstress_meta.json")
-    if p2.exists():
-        try:
-            return json.loads(p2.read_text(encoding="utf-8"))
-        except Exception:
-            pass
-
-    return {}
-
 # =========================
 # Session state
 # =========================
@@ -566,9 +216,12 @@ st.session_state.setdefault("dev_file_bytes",b"")
 st.session_state.setdefault("dev_file_loaded",False)
 st.session_state.setdefault("dev_preview",False)
 st.session_state.setdefault("show_preview_modal", False)
+st.session_state.setdefault("model_loaded", False)
+st.session_state.setdefault("model_obj", None)
+st.session_state.setdefault("meta_dict", {})
 
 # =========================
-# Sidebar branding
+# Sidebar: branding + model upload
 # =========================
 st.sidebar.markdown(f"""
     <div class="centered-container">
@@ -578,45 +231,36 @@ st.sidebar.markdown(f"""
     </div>
 """, unsafe_allow_html=True)
 
-def sticky_header(title, message):
-    st.markdown(
-        f"""
-        <style>
-        .sticky-container {{
-            position: sticky; top: 0; background-color: white; z-index: 100;
-            padding-top: 10px; padding-bottom: 10px; border-bottom: 1px solid #eee;
-        }}
-        </style>
-        <div class="sticky-container">
-            <h3>{title}</h3>
-            <p>{message}</p>
-        </div>
-        """,
-        unsafe_allow_html=True
-    )
+with st.sidebar.expander("① Load model (upload)", expanded=True):
+    up_model = st.file_uploader("Model file (.joblib)", type=["joblib","pkl"], key="mdl_up")
+    up_meta  = st.file_uploader("Meta file (.json)", type=["json"], key="meta_up")
+    load_btn = st.button("Load model", type="primary")
 
-# =========================
-# Load model + meta
-# =========================
-try:
-    model = load_model_from_hub_or_local()
-except Exception as e:
-    st.error(
-        "Failed to load model.\n"
-        "• Ensure HF_REPO_ID, MODEL_FILE are correct and the repo/file exists.\n"
-        "• If the repo is private, set AUTH_TOKEN (Read).\n"
-        "• Or upload /mnt/data/minstress_model.joblib for testing.\n\n"
-        f"Details: {e}"
-    )
-    st.stop()
+    if load_btn:
+        if not up_model:
+            st.error("Please upload the model .joblib file.")
+            st.stop()
+        try:
+            st.session_state.model_obj = joblib.load(io.BytesIO(up_model.getvalue()))
+            st.session_state.model_loaded = True
+        except Exception as e:
+            st.error(f"Failed to load model: {e}")
+            st.stop()
+
+        if up_meta:
+            try:
+                st.session_state.meta_dict = json.loads(up_meta.getvalue().decode("utf-8"))
+            except Exception as e:
+                st.warning(f"Could not parse meta.json: {e}")
+                st.session_state.meta_dict = {}
+        else:
+            st.warning("No meta.json uploaded — using app defaults.")
+            st.session_state.meta_dict = {}
 
-meta = {}
-ALIASES = None
-try:
-    meta = load_meta_from_hub_or_local()
-except Exception as e:
-    st.warning(f"Could not load meta.json: {e}")
+        st.success("Model loaded in memory ✓")
 
+# Apply meta (if provided)
+meta = st.session_state.meta_dict
 if meta:
     FEATURES   = meta.get("features", FEATURES)
     TARGET     = meta.get("target", TARGET)
@@ -634,18 +278,46 @@ if meta:
             msg.append(f"scikit-learn {mv['scikit_learn']} expected, running {_skl.__version__}")
         if msg:
             st.warning("Environment mismatch: " + " | ".join(msg))
+else:
+    ALIASES = None
+
+# Guard: require model first
+if not st.session_state.model_loaded:
+    st.header("Welcome!")
+    st.info("Upload your **model** (.joblib) and optional **meta.json** in the left sidebar, then click **Load model**.")
+    st.stop()
+
+# Keep a short alias
+model = st.session_state.model_obj
+
+# =========================
+# Sticky header helper
+# =========================
+def sticky_header(title, message):
+    st.markdown(
+        f"""
+        <style>
+        .sticky-container {{
+            position: sticky; top: 0; background-color: white; z-index: 100;
+            padding-top: 10px; padding-bottom: 10px; border-bottom: 1px solid #eee;
+        }}
+        </style>
+        <div class="sticky-container">
+            <h3>{title}</h3>
+            <p>{message}</p>
+        </div>
+        """,
+        unsafe_allow_html=True
+    )
 
 # =========================
 # INTRO
 # =========================
 if st.session_state.app_step == "intro":
-    st.header("Welcome!")
-    st.markdown(f"This software is developed by *Smart Thinking AI-Solutions Team* to estimate **Minimum Horizontal Stress** ({UNITS}) from drilling/offset data.")
-    st.subheader("How It Works")
+    st.header("Model ready ✓")
     st.markdown(
-        "1) **Upload your data to build the case and preview the model performance.**  \n"
-        "2) Click **Run Model** to compute metrics and plots.  \n"
-        "3) **Proceed to Validation** (with actual) or **Proceed to Prediction** (no actual)."
+        f"This software estimates **Minimum Horizontal Stress** ({UNITS}). "
+        "Now build a case, validate, or predict."
     )
     if st.button("Start Showcase", type="primary"):
         st.session_state.app_step = "dev"; st.rerun()
@@ -660,8 +332,8 @@ def _find_sheet(book, names):
     return None
 
 if st.session_state.app_step == "dev":
-    st.sidebar.header("Case Building")
-    up = st.sidebar.file_uploader("Upload Your Data File (Train/Test sheets)", type=["xlsx","xls"])
+    st.sidebar.header("② Case Building")
+    up = st.sidebar.file_uploader("Upload Train/Test Excel", type=["xlsx","xls"])
     if up is not None:
         st.session_state.dev_file_bytes = up.getvalue()
         st.session_state.dev_file_name = up.name
@@ -751,13 +423,61 @@ if st.session_state.app_step == "dev":
             with tab1: _dev_block(st.session_state.results["Train"], st.session_state.results["m_train"])
         if "Test" in st.session_state.results:
             with tab2: _dev_block(st.session_state.results["Test"],  st.session_state.results["m_test"])
-        render_export_button(phase_key="dev")
+        # Export
+        st.divider()
+        st.markdown("### Export to Excel")
+        options = ["Training","Training_Metrics","Training_Summary","Testing","Testing_Metrics","Testing_Summary","Info"]
+        selected = st.multiselect("Sheets to include", options=options, default=[])
+        def _summary_table(df: pd.DataFrame, cols: list[str]) -> pd.DataFrame:
+            cols = [c for c in cols if c in df.columns]
+            if not cols: return pd.DataFrame()
+            tbl = (df[cols].agg(['min','max','mean','std'])
+                   .T.rename(columns={"min":"Min","max":"Max","mean":"Mean","std":"Std"})
+                   .reset_index(names="Field"))
+            return _round_numeric(tbl, 3)
+        def build_export(selected: list[str]) -> tuple[bytes|None, str|None]:
+            res = st.session_state.get("results", {})
+            if not res: return None, None
+            sheets, order = {}, []
+            def _add(n, d): 
+                if isinstance(d, pd.DataFrame) and not d.empty: sheets[n]=_round_numeric(d,3); order.append(n)
+            if "Training" in selected and "Train" in res: _add("Training", res["Train"])
+            if "Training_Metrics" in selected and res.get("m_train"): _add("Training_Metrics", pd.DataFrame([res["m_train"]]))
+            if "Training_Summary" in selected and "Train" in res:
+                tr_cols = FEATURES + [c for c in [TARGET, PRED_COL] if c in res["Train"].columns]
+                _add("Training_Summary", _summary_table(res["Train"], tr_cols))
+            if "Testing" in selected and "Test" in res: _add("Testing", res["Test"])
+            if "Testing_Metrics" in selected and res.get("m_test"): _add("Testing_Metrics", pd.DataFrame([res["m_test"]]))
+            if "Testing_Summary" in selected and "Test" in res:
+                te_cols = FEATURES + [c for c in [TARGET, PRED_COL] if c in res["Test"].columns]
+                _add("Testing_Summary", _summary_table(res["Test"], te_cols))
+            if "Info" in selected:
+                info = pd.DataFrame([
+                    {"Key":"AppName","Value":APP_NAME},
+                    {"Key":"Tagline","Value":TAGLINE},
+                    {"Key":"Target","Value":TARGET},
+                    {"Key":"PredColumn","Value":PRED_COL},
+                    {"Key":"Features","Value":", ".join(FEATURES)},
+                    {"Key":"ExportedAt","Value":datetime.now().strftime("%Y-%m-%d %H:%M:%S")},
+                ])
+                _add("Info", info)
+            if not order: return None, None
+            bio = io.BytesIO()
+            with pd.ExcelWriter(bio, engine=_excel_engine()) as w:
+                for name in order:
+                    df = sheets[name]; df.to_excel(w, sheet_name=_excel_safe_name(name), index=False)
+            bio.seek(0)
+            return bio.getvalue(), f"MinStress_Export_{datetime.now().strftime('%Y%m%d_%H%M%S')}.xlsx"
+        data, fname = build_export(selected)
+        st.download_button("⬇️ Export Excel", data=(data or b""), file_name=(fname or "MinStress_Export.xlsx"),
+                           mime="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
+                           disabled=(data is None))
 
 # =========================
 # VALIDATION (with actual)
 # =========================
 if st.session_state.app_step == "validate":
-    st.sidebar.header("Validate the Model")
+    st.sidebar.header("③ Validate the Model")
     up = st.sidebar.file_uploader("Upload Validation Excel", type=["xlsx","xls"])
     if up is not None:
         book = read_book_bytes(up.getvalue())
@@ -777,11 +497,9 @@ if st.session_state.app_step == "validate":
         names = list(book.keys())
         name = next((s for s in names if s.lower() in ("validation","validate","validation2","val","val2")), names[0])
         df0 = _normalize_columns(book[name].copy(), FEATURES, TARGET, ALIASES)
-
         act_col = ACTUAL_COL if (ACTUAL_COL and ACTUAL_COL in df0.columns) else TARGET
         if not ensure_cols(df0, FEATURES+[act_col]):
             st.markdown('<div class="st-message-box st-error">Missing required columns.</div>', unsafe_allow_html=True); st.stop()
-
         df = df0.copy()
         df[PRED_COL] = _inv_transform(model.predict(_make_X(df0, FEATURES)), TRANSFORM)
         st.session_state.results["Validate"] = df
@@ -828,19 +546,23 @@ if st.session_state.app_step == "validate":
                                         st.session_state.results["Validate"][PRED_COL]),
                       use_container_width=False)
 
-        render_export_button(phase_key="validate")
-
-        sv = st.session_state.results["sv_val"]
-        if sv["oor"] > 0: st.markdown('<div class="st-message-box st-warning">Some inputs fall outside **training min–max** ranges.</div>', unsafe_allow_html=True)
-        if st.session_state.results["oor_tbl"] is not None:
-            st.write("*Out-of-range rows (vs. Training min–max):*")
-            df_centered_rounded(st.session_state.results["oor_tbl"])
+        # Export button
+        st.divider()
+        val_tbl = st.session_state.results["Validate"]
+        bio = io.BytesIO()
+        with pd.ExcelWriter(bio, engine=_excel_engine()) as w:
+            val_tbl.to_excel(w, sheet_name="Validation", index=False)
+            pd.DataFrame([m]).to_excel(w, sheet_name="Validation_Metrics", index=False)
+        bio.seek(0)
+        st.download_button("⬇️ Export Excel", data=bio.getvalue(),
+                           file_name=f"Validation_{datetime.now().strftime('%Y%m%d_%H%M%S')}.xlsx",
+                           mime="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet")
 
 # =========================
 # PREDICTION (no actual)
 # =========================
 if st.session_state.app_step == "predict":
-    st.sidebar.header("Prediction (No Actual)")
+    st.sidebar.header("④ Prediction (No Actual)")
     up = st.sidebar.file_uploader("Upload Prediction Excel", type=["xlsx","xls"])
     if up is not None:
         book = read_book_bytes(up.getvalue())
@@ -887,11 +609,10 @@ if st.session_state.app_step == "predict":
             })
             st.markdown('<div class="st-message-box st-success">Predictions ready ✓</div>', unsafe_allow_html=True)
             df_centered_rounded(table, hide_index=True)
-            st.caption("**★ OOR** = %% of rows with input features outside the training min–max range.")
+            st.caption("**★ OOR** = % of rows with input features outside the training min–max range.")
         with col_right:
             st.plotly_chart(track_plot(df, include_actual=False),
                             use_container_width=False, config={"displayModeBar": False, "scrollZoom": True})
-        render_export_button(phase_key="predict")
 
 # =========================
 # Preview modal
@@ -914,20 +635,33 @@ if st.session_state.show_preview_modal:
                     df = _normalize_columns(book_to_preview[name], FEATURES, TARGET, ALIASES)
                     t1, t2 = st.tabs(["Tracks", "Summary"])
                     with t1:
-                        st.pyplot(preview_tracks(df, FEATURES), use_container_width=True)
+                        # small quick-look plot of the features
+                        cols = [c for c in FEATURES if c in df.columns]
+                        if not cols:
+                            st.info("No feature columns to preview.")
+                        else:
+                            idx = np.arange(1, len(df)+1)
+                            fig, axes = plt.subplots(1, len(cols), figsize=(2.4*len(cols), 7.0), sharey=True, dpi=100)
+                            if len(cols)==1: axes=[axes]
+                            for ax, col in zip(axes, cols):
+                                x = pd.to_numeric(df[col], errors="coerce")
+                                ax.plot(x, idx, '-', lw=1.6)
+                                ax.set_xlabel(col); ax.xaxis.set_label_position('top'); ax.xaxis.tick_top()
+                                ax.set_ylim(idx.max(), idx.min()); ax.grid(True, linestyle=":", alpha=0.3)
+                            fig.tight_layout()
+                            st.pyplot(fig, use_container_width=True)
                     with t2:
-                        feat_present = [c for c in FEATURES if c in df.columns]
-                        if not feat_present:
+                        cols = [c for c in FEATURES if c in df.columns]
+                        if not cols:
                             st.info("No feature columns found to summarize.")
                         else:
                             tbl = (
-                                df[feat_present]
+                                df[cols]
                                   .agg(['min','max','mean','std'])
                                   .T.rename(columns={"min":"Min","max":"Max","mean":"Mean","std":"Std"})
                                   .reset_index(names="Feature")
                             )
                             df_centered_rounded(tbl)
-
     st.session_state.show_preview_modal = False
 
 # =========================