Update app.py

app.py

@@ -67,17 +67,106 @@ def load_bible_names():
 ssa_names_df, ssa_names_aggregated_df = load_ssa_names()
 bible_names_df = load_bible_names()
 
+
+# -----------------------
+# filtering and picking 
+# -----------------------
+
+def get_normal_and_bible(
+    ssa_names_df,
+    bible_names_df,
+    min_length_ssa=1,
+    max_length_ssa=None,
+    min_length_bible=1,
+    max_length_bible=None,
+    ssa_popularity_percentile=None,
+    sex=None,
+    forbidden_names=None,
+):
+    filtered_ssa_df = ssa_names_df
+
+    if ssa_popularity_percentile is not None:
+        low, high = ssa_popularity_percentile
+        name_counts = (
+            ssa_names_df.groupby("name", as_index=False)["count"]
+            .sum()
+            .sort_values(by="count", ascending=True)
+        )
+        total = len(name_counts)
+        selected_names = name_counts.iloc[int(low * total):int(high * total)]["name"]
+        filtered_ssa_df = ssa_names_df[ssa_names_df["name"].isin(selected_names)]
+
+    if sex:
+        filtered_ssa_df = filtered_ssa_df[filtered_ssa_df["sex"] == sex]
+        bible_names_df = bible_names_df[bible_names_df["sex"] == sex]
+
+    if forbidden_names:
+        filtered_ssa_df = filtered_ssa_df[~filtered_ssa_df["name"].isin(forbidden_names)]
+        bible_names_df = bible_names_df[~bible_names_df["english_label"].isin(forbidden_names)]
+
+    ssa_mask = filtered_ssa_df["name"].str.len() >= min_length_ssa
+    if max_length_ssa is not None:
+        ssa_mask &= filtered_ssa_df["name"].str.len() <= max_length_ssa
+    normal_names = filtered_ssa_df[ssa_mask]["name"].unique().tolist()
+
+    bible_mask = bible_names_df["english_label"].str.len() >= min_length_bible
+    if max_length_bible is not None:
+        bible_mask &= bible_names_df["english_label"].str.len() <= max_length_bible
+    bible_names = bible_names_df[bible_mask]["english_label"].unique().tolist()
+
+    if not normal_names or not bible_names:
+        raise ValueError("No names found with given constraints")
+
+    return random.choice(normal_names), random.choice(bible_names)
+
+def generate_names(n, sex, min_len, max_len, min_bible_len, max_bible_len, pop_low, pop_high,last="Smith"):
+    names = []
+    for _ in range(n):
+        try:
+            normal, bible = get_normal_and_bible(
+                ssa_names_aggregated_df,
+                bible_names_personlabel_df,
+                min_length_ssa=min_len,
+                max_length_ssa=max_len,
+                min_length_bible=min_bible_len,
+                max_length_bible=max_bible_len,
+                ssa_popularity_percentile=(pop_low, pop_high),
+                sex=sex if sex in {"M", "F"} else None
+            )
+            names.append(f"{bible} {normal} {last")
+        except Exception as e:
+            names.append(f"[Error: {e}]")
+    return "\n".join(names)
+
 # ------------------
 # Gradio Interface
 # ------------------
 
 with gr.Blocks() as demo:
-    gr.Markdown("## Bible Names & SSA Names Datasets")
+    gr.Markdown("## 📜 Bible + SSA Name Generator")
+
+    with gr.Row():
+        n_slider = gr.Slider(1, 100, value=10, label="Number of names")
+        sex_choice = gr.Radio(["M", "F", "Any"], label="Sex", value="Any")
 
-    with gr.Tab("Bible Names"):
-        gr.Dataframe(bible_names_df.head(100), label="Bible Names (first 100 rows)")
+    with gr.Row():
+        pop_range = gr.RangeSlider(0.0, 1.0, value=[0.0, 1.0], step=0.01, label="Popularity Percentile Range (SSA)")
 
-    with gr.Tab("SSA Names (Aggregated)"):
-        gr.Dataframe(ssa_names_aggregated_df.head(100), label="Top SSA Names (first 100 rows)")
+    with gr.Row():
+        ssa_len = gr.Slider(1, 15, value=1, label="SSA Name Length (min)")
+        ssa_max_len = gr.Slider(1, 15, value=10, label="SSA Name Length (max)")
+
+    with gr.Row():
+        bible_len = gr.Slider(1, 15, value=4, label="Bible Name Length (min)")
+        bible_max_len = gr.Slider(1, 15, value=10, label="Bible Name Length (max)")
+
+    generate_btn = gr.Button("Generate Names")
+    output_box = gr.Textbox(label="Generated Names", lines=15)
+
+    generate_btn.click(
+        fn=generate_names,
+        inputs=[n_slider, sex_choice, ssa_len, ssa_max_len, bible_len, bible_max_len, pop_range],
+        outputs=output_box
+    )
 
 demo.launch()