Update app.py

app.py

@@ -13,7 +13,7 @@ from sklearn.neighbors import NearestNeighbors
 import gradio as gr
 
 # ============= Config =============
-USE_EMBEDDINGS = False              # keep False for now (baseline). Can upgrade later.
+USE_EMBEDDINGS = False
 MAX_ROWS = int(os.getenv("MAX_ROWS", "5000"))
 DEFAULT_OCCASIONS = "birthday, thank_you, housewarming"
 
@@ -43,42 +43,31 @@ def _to_price_usd(x):
     except: return np.nan
 
 def _infer_age_from_category(cat: str) -> str:
-    """Very simple heuristic: map Amazon categories to an age_range tag."""
     s = (cat or "").lower()
-    if any(k in s for k in ["baby", "toddler", "infant"]):
-        return "kids"
-    if "toys & games" in s or "board games" in s or "toy" in s:
-        return "kids"
-    if any(k in s for k in ["teen", "ya", "young adult"]):
-        return "teens"
-    # otherwise
+    if any(k in s for k in ["baby", "toddler", "infant"]): return "kids"
+    if "toys & games" in s or "board games" in s or "toy" in s: return "kids"
+    if any(k in s for k in ["teen", "ya", "young adult"]): return "teens"
     return "any"
 
 def map_amazon_to_schema(df_raw: pd.DataFrame) -> pd.DataFrame:
     cols = {c.lower().strip(): c for c in df_raw.columns}
     get = lambda key: df_raw.get(cols.get(key, ""), "")
-
     out = pd.DataFrame({
         "name": get("product name"),
         "short_desc": get("description"),
         "tags": get("category"),
         "price_usd": get("selling price").map(_to_price_usd) if "selling price" in cols else np.nan,
-        "age_range": "",   # fill by inference below
+        "age_range": "",
         "gender_tags": "any",
         "occasion_tags": DEFAULT_OCCASIONS,
         "persona_fit": get("category"),
         "image_url": get("image") if "image" in cols else "",
     })
-
-    # clean
-    out["name"] = out["name"].astype(str).str.strip().str.slice(0, 120)
-    out["short_desc"] = out["short_desc"].astype(str).str.strip().str.slice(0, 400)
+    out["name"] = out["name"].astype(str).str.strip().str.slice(0,120)
+    out["short_desc"] = out["short_desc"].astype(str).str.strip().str.slice(0,400)
     out["tags"] = out["tags"].astype(str).str.replace("|", ", ").str.lower()
     out["persona_fit"] = out["persona_fit"].astype(str).str.lower()
-
-    # age inference
     out["age_range"] = out["tags"].map(_infer_age_from_category).fillna("any")
-
     return out
 
 def build_doc(row: pd.Series) -> str:
@@ -97,7 +86,6 @@ def load_catalog() -> pd.DataFrame:
         ds = load_dataset("ckandemir/amazon-products", split="train")
         raw = ds.to_pandas()
     except Exception:
-        # tiny fallback so the app doesn't crash offline
         raw = pd.DataFrame({
             "Product Name": ["Wireless Earbuds", "Coffee Sampler", "Strategy Board Game"],
             "Description": [
@@ -144,11 +132,11 @@ def filter_business(df: pd.DataFrame, budget_min=None, budget_max=None,
     if occasion:
         m &= _contains_ci(df["occasion_tags"], occasion)
     if age_range and age_range != "any":
-        # keep items explicitly tagged as this age OR 'any'
         m &= (df["age_range"].fillna("any").isin([age_range, "any"]))
     return df[m]
 
 def recommend_topk(profile: Dict, k: int=3) -> pd.DataFrame:
+    """Global kNN → filter to the business subset (fixes index mismatch)."""
     q = profile_to_query(profile)
     q_vec = _vectorizer.transform([q])
 
@@ -162,24 +150,37 @@ def recommend_topk(profile: Dict, k: int=3) -> pd.DataFrame:
     if df_f.empty:
         df_f = CATALOG
 
-    idx = df_f.index.values
-    dists, inds = _nn.kneighbors(q_vec, n_neighbors=min(max(k*4,k), len(df_f)))
-    cand_idx = idx[inds[0]]
+    # Search on the global index, then keep only rows inside df_f
+    n_cand = min(max(k*50, k), len(CATALOG))
+    dists, inds = _nn.kneighbors(q_vec, n_neighbors=n_cand)
+    cand_global = inds[0]              # indices in CATALOG
     d = dists[0]
-    order = np.argsort(d)
-    cand_idx = cand_idx[order]
-    d = d[order]
-
+    order = np.argsort(d)              # ascending distance
     seen, picks = set(), []
-    for ci, dist in zip(cand_idx, d):
-        nm = CATALOG.loc[ci, "name"]
-        if nm in seen: continue
+    for gi in cand_global[order]:
+        if gi not in df_f.index:       # keep only filtered subset
+            continue
+        nm = CATALOG.loc[gi, "name"]
+        if nm in seen:
+            continue
         seen.add(nm)
-        picks.append((ci, 1 - float(dist)))
-        if len(picks) >= k: break
-
-    res = CATALOG.loc[[ci for ci,_ in picks]].copy()
-    res["similarity"] = [sim for _,sim in picks]
+        # similarity = 1 - distance
+        sim = 1 - float(_nn.kneighbors_graph(q_vec, n_neighbors=1, mode="distance")[0, gi]) if False else 1.0
+        # we already have distances in d; recompute sim from them using same order index:
+        # get distance for this gi:
+        # (for simplicity we just set sim to 1 - current min distance; not critical for UI ranking)
+        picks.append((gi, None))
+        if len(picks) >= k:
+            break
+
+    if not picks:
+        return df_f.head(k).assign(similarity=np.nan)[["name","short_desc","price_usd","occasion_tags","persona_fit","age_range","image_url","similarity"]]
+
+    sel = [gi for gi,_ in picks]
+    res = CATALOG.loc[sel].copy()
+    # compute similarity from the original distances vector for display
+    gi_to_dist = {int(gi): float(dist) for gi, dist in zip(cand_global, d)}
+    res["similarity"] = [1.0 - gi_to_dist.get(int(gi), 0.0) for gi in sel]
     return res[["name","short_desc","price_usd","occasion_tags","persona_fit","age_range","image_url","similarity"]]
 
 # ============= Synthetic item + message =============
@@ -200,15 +201,12 @@ def generate_item(profile: Dict) -> Dict:
     else:
         base_name = f"{core.title()} starter bundle"
         base_desc = f"A curated set to kickstart their {core} passion."
-
-    # small age tweak
     if age == "kids":
         base_desc += " Suitable for kids with safe, age-appropriate materials."
     elif age == "teens":
         base_desc += " Trendy pick that suits young enthusiasts."
     elif age == "senior":
         base_desc += " Comfortable and easy to use."
-
     price = float(np.clip(float(budget), 10, 250))
     return {
         "name": f"{base_name} ({occasion})",
@@ -230,26 +228,29 @@ def generate_message(profile: Dict) -> str:
 
 # ============= Gradio UI (GIfty) =============
 EXAMPLES = [
-    [["reading","travel","aesthetic"], "birthday", 20, 60, "Noa", "adult", "warm and friendly"],
-    [["coffee","home","practical"], "housewarming", 25, 45, "Daniel", "adult", "warm"],
-    [["tech","photography" if "photography" in INTEREST_OPTIONS else "digital"], "birthday", 30, 120, "Omer", "teen (13–17)", "fun"],
+    [["reading","travel","aesthetic"], "birthday", [20, 60], "Noa", "adult (18–64)", "warm and friendly"],
+    [["coffee","home","practical"], "housewarming", [25, 45], "Daniel", "adult (18–64)", "warm"],
+    [["tech","digital"], "birthday", [30, 120], "Omer", "teen (13–17)", "fun"],
 ]
 
-def ui_predict(interests_list: List[str], occasion: str, budget_min: float, budget_max: float,
-               recipient_name: str, age_label: str, tone: str):
+def ui_predict(interests_list: List[str], occasion: str, budget_range, recipient_name: str, age_label: str, tone: str):
     try:
-        age_range = AGE_OPTIONS.get(age_label, "any")
-        # Ensure min <= max
-        if budget_min is not None and budget_max is not None and budget_min > budget_max:
+        # budget_range is a tuple/list: (min, max)
+        if isinstance(budget_range, (list, tuple)) and len(budget_range) == 2:
+            budget_min, budget_max = float(budget_range[0]), float(budget_range[1])
+        else:
+            budget_min, budget_max = 20.0, 60.0
+        if budget_min > budget_max:
             budget_min, budget_max = budget_max, budget_min
 
+        age_range = AGE_OPTIONS.get(age_label, "any")
         profile = {
             "recipient_name": recipient_name or "Friend",
             "interests": interests_list or [],
             "occasion": occasion or "birthday",
-            "budget_min": float(budget_min) if budget_min is not None else None,
-            "budget_max": float(budget_max) if budget_max is not None else None,
-            "budget_usd": float(budget_max) if budget_max is not None else 50.0,
+            "budget_min": budget_min,
+            "budget_max": budget_max,
+            "budget_usd": budget_max,
             "age_range": age_range,
             "tone": tone or "warm and friendly",
         }
@@ -262,7 +263,6 @@ def ui_predict(interests_list: List[str], occasion: str, budget_min: float, budg
         gen_md = f"**{gen['name']}**\n\n{gen['short_desc']}\n\n~${gen['price_usd']:.0f}"
         return top3_md, gen_md, msg
     except Exception as e:
-        # Return error message in the UI instead of "nothing happens"
         return f":warning: Error: {e}", "", ""
 
 with gr.Blocks() as demo:
@@ -276,9 +276,10 @@ with gr.Blocks() as demo:
     with gr.Row():
         occasion = gr.Dropdown(label="Occasion", choices=OCCASION_OPTIONS, value="birthday")
         age = gr.Dropdown(label="Age group", choices=list(AGE_OPTIONS.keys()), value="adult (18–64)")
-    with gr.Row():
-        budget_min = gr.Slider(label="Budget min (USD)", minimum=5, maximum=300, step=1, value=20)
-        budget_max = gr.Slider(label="Budget max (USD)", minimum=10, maximum=500, step=1, value=60)
+
+    # Range slider for budget (two handles)
+    budget = gr.Slider(label="Budget (USD)", minimum=5, maximum=500, step=1, value=(20, 60))
+
     with gr.Row():
         recipient_name = gr.Textbox(label="Recipient name", value="Noa")
         tone = gr.Textbox(label="Message tone", value="warm and friendly")
@@ -290,16 +291,15 @@ with gr.Blocks() as demo:
 
     gr.Examples(
         EXAMPLES,
-        [interests, occasion, budget_min, budget_max, recipient_name, age, tone],
+        [interests, occasion, budget, recipient_name, age, tone],
         label="Quick examples",
     )
 
     go.click(
         ui_predict,
-        [interests, occasion, budget_min, budget_max, recipient_name, age, tone],
+        [interests, occasion, budget, recipient_name, age, tone],
         [out_top3, out_gen, out_msg]
     )
 
-# For Spaces
 if __name__ == "__main__":
     demo.launch()