Update app.py

app.py

@@ -1,8 +1,8 @@
 # ================================================================
 # Self-Sensing Concrete Assistant — Predictor (XGB) + Hybrid RAG
-# - Predictor tab: identical behavior (kept) + * marks for required fields
-# - Literature tab: Hybrid RAG; LLM runs silently when available
-# - UX: no visible "LLM & Controls" window; prediction=0.0 if incomplete/error
+# - Predictor tab: required fields marked with *
+# - Prediction fixed: NA is accepted for required categoricals
+# - RAG page-extraction regex fixed for [[PAGE=...]]
 # ================================================================
 
 # ---------------------- Runtime flags (HF-safe) ----------------------
@@ -49,7 +49,7 @@ LLM_AVAILABLE = (OPENAI_API_KEY is not None and OPENAI_API_KEY.strip() != "" and
 # ========================= Predictor (kept) =========================
 CF_COL     = "Conductive Filler Conc. (wt%)"
 TARGET_COL = "Stress GF (MPa-1)"
-CANON_NA   = "NA"  # <-- canonical placeholder for categoricals
+CANON_NA   = "NA"  # canonical placeholder for categoricals
 
 MAIN_VARIABLES = [
     "Filler 1 Type",
@@ -113,7 +113,7 @@ OPTIONAL_FIELDS = {
     "Filler 2 Dimensionality",
 }
 
-# Only these fields are required
+# Required fields (as you specified earlier)
 REQUIRED_FIELDS = {
     "Filler 1 Type",
     "Filler 1 Diameter (µm)",
@@ -177,12 +177,16 @@ def _coerce_to_row(form_dict: dict) -> pd.DataFrame:
         elif col in CATEGORICAL_COLS:
             row[col] = _canon_cat(v)
         else:
-            # non-numeric, non-categorical (free text) — keep trimmed, but not empty
             s = str(v).strip() if v is not None else ""
             row[col] = s if s else CANON_NA
     return pd.DataFrame([row], columns=MAIN_VARIABLES)
 
 def _is_complete(form_dict: dict) -> bool:
+    """
+    FIX: For required *categoricals*, NA counts as 'provided' (acceptable),
+    so users aren't blocked when NA is a legitimate choice.
+    Numeric required fields must be non-NaN.
+    """
     for col in REQUIRED_FIELDS:
         v = form_dict.get(col, None)
         if col in NUMERIC_COLS:
@@ -192,8 +196,10 @@ def _is_complete(form_dict: dict) -> bool:
             except Exception:
                 return False
         else:
+            # Required categoricals/text: accept any non-empty after canonicalization,
+            # and accept CANON_NA as "provided".
             s = _canon_cat(v)
-            if s == CANON_NA:
+            if s == "" or s is None:
                 return False
     return True
 
@@ -212,7 +218,7 @@ def _align_columns_to_model(df: pd.DataFrame, mdl) -> pd.DataFrame:
     return df
 
 def predict_fn(**kwargs):
-    # Return a NUMBER always (0.0 on incomplete or any error) to keep gr.Number happy
+    # Keep your contract: 0.0 if incomplete or on error
     if not _is_complete(kwargs):
         return 0.0
     mdl = _load_model_or_error()
@@ -221,13 +227,14 @@ def predict_fn(**kwargs):
     X_new = _coerce_to_row(kwargs)
     X_new = _align_columns_to_model(X_new, mdl)
     try:
-        y_raw = mdl.predict(X_new)               # model may predict log1p(target) or original scale
-        # If your model was trained on log1p, set mdl.target_is_log1p_ = True before saving.
+        y_raw = mdl.predict(X_new)  # log1p or original scale depending on training
         if getattr(mdl, "target_is_log1p_", False):
             y = np.expm1(y_raw)
         else:
             y = y_raw
-        y = float(np.maximum(np.asarray(y).ravel()[0], 0.0))
+        y = float(np.asarray(y).ravel()[0])
+        if y < 0:
+            y = 0.0
         return y
     except Exception as e:
         print(f"[Predict] {e}")
@@ -404,6 +411,7 @@ bm25 = BM25Okapi(bm25_tokens) if (BM25Okapi is not None and bm25_tokens is not N
 st_query_model = _safe_init_st_model(os.getenv("EMB_MODEL_NAME", "sentence-transformers/all-MiniLM-L6-v2"))
 
 def _extract_page(text_chunk: str) -> str:
+    # FIXED: proper brackets; matches [[PAGE=123]]
     m = list(re.finditer(r"\[\[PAGE=(\d+)\]\]", text_chunk or ""))
     return (m[-1].group(1) if m else "?")
 
@@ -692,7 +700,6 @@ with gr.Blocks(css=CSS, theme=theme, fill_height=True) as demo:
             with gr.Row():
                 with gr.Column(scale=7):
                     with gr.Accordion("Primary conductive filler", open=True, elem_classes=["card"]):
-                        # * marks on required inputs only
                         f1_type = gr.Textbox(label="Filler 1 Type *", placeholder="e.g., CNT, Graphite, Steel fiber")
                         f1_diam = gr.Number(label="Filler 1 Diameter (µm) *")
                         f1_len  = gr.Number(label="Filler 1 Length (mm) *")
@@ -776,11 +783,11 @@ with gr.Blocks(css=CSS, theme=theme, fill_height=True) as demo:
                 w_bm25  = gr.Slider(0.0, 1.0, value=W_BM25_DEFAULT,  step=0.05, label="BM25 weight")
                 w_emb   = gr.Slider(0.0, 1.0, value=(0.0 if not USE_DENSE else 0.40), step=0.05, label="Dense weight (set 0 if disabled)")
 
-            # ---- Hidden states for LLM behavior (no visible controls) ----
-            state_use_llm     = gr.State(LLM_AVAILABLE)   # True when key present; else False
+            # Hidden states (unchanged)
+            state_use_llm     = gr.State(LLM_AVAILABLE)
             state_model_name  = gr.State(os.getenv("OPENAI_MODEL", OPENAI_MODEL))
             state_temperature = gr.State(0.2)
-            state_strict      = gr.State(False)           # hidden: default to not-strict
+            state_strict      = gr.State(False)
 
             gr.ChatInterface(
                 fn=rag_chat_fn,