Upload folder using huggingface_hub

app.py

@@ -1,4 +1,5 @@
 import json
+import math
 import os
 import re
 import glob
@@ -736,6 +737,7 @@ def clustering_tab():
                 ),
             ], width=4),
         ], className="mb-2"),
+        html.Div(id="diversity-cards", className="mb-2"),
         dbc.Row([dbc.Col(html.Div(id="plot-status", className="text-danger small mb-1"))]),
         dbc.Row([
             dbc.Col(
@@ -2300,9 +2302,69 @@ def populate_cross_filter_values(cross_field, model, min_val, threshold):
     return options, False, None
 
 
+def _compute_diversity_cards(field_df):
+    """Return a dbc.Row of diversity metric cards from a filtered field DataFrame."""
+    n_total = len(field_df)
+    if n_total == 0:
+        return html.Div()
+
+    clustered = field_df[field_df["cluster_id"] != -1]
+    n_clustered = len(clustered)
+    coverage = n_clustered / n_total if n_total > 0 else 0.0
+
+    if n_clustered == 0:
+        return _diversity_row(n_total, 0, 0, 0.0, 0.0, coverage, 0.0)
+
+    sizes = clustered["cluster_id"].value_counts()
+    n_clusters = len(sizes)
+    p = sizes / n_clustered
+    entropy = -sum(pi * math.log(pi) for pi in p if pi > 0)
+    effective_n = math.exp(entropy) if entropy > 0 else 1.0
+    dominance = sizes.iloc[0] / n_clustered  # sizes is sorted descending
+
+    return _diversity_row(n_total, n_clustered, n_clusters, entropy, effective_n, coverage, dominance)
+
+
+def _diversity_row(n_total, n_clustered, n_clusters, entropy, effective_n, coverage, dominance):
+    def _card(label, value, tooltip, color="#1a3a5c"):
+        return dbc.Col(
+            dbc.Card(
+                dbc.CardBody([
+                    html.Div(label, className="text-muted mb-1",
+                             style={"fontSize": "11px", "textTransform": "uppercase",
+                                    "letterSpacing": "0.05em"}),
+                    html.Div(value, className="fw-bold",
+                             style={"fontSize": "20px", "color": color}),
+                    html.Div(tooltip, className="text-muted",
+                             style={"fontSize": "10px", "lineHeight": "1.3"}),
+                ], className="p-2 text-center"),
+                className="border-0 shadow-sm h-100",
+            ),
+            xs=6, sm=4, md=2,
+        )
+
+    return dbc.Row([
+        _card("Entries",       f"{n_total:,}",
+              "total in field after filters"),
+        _card("Clusters",      f"{n_clusters:,}",
+              "distinct clusters (excl. noise)"),
+        _card("Coverage",      f"{coverage:.1%}",
+              "assigned to a cluster (not noise)",
+              "#198754" if coverage >= 0.7 else "#dc3545"),
+        _card("Shannon H",     f"{entropy:.2f}",
+              "entropy of cluster-size distribution"),
+        _card("Effective N",   f"{effective_n:.1f}",
+              "exp(H) — diversity-adjusted cluster count"),
+        _card("Dominance",     f"{dominance:.1%}",
+              "entries in the single largest cluster",
+              "#dc3545" if dominance >= 0.5 else "#1a3a5c"),
+    ], className="g-2")
+
+
 @app.callback(
     Output("cluster-graph", "figure"),
     Output("plot-status", "children"),
+    Output("diversity-cards", "children"),
     Input("dd-model", "value"),
     Input("dd-min", "value"),
     Input("dd-threshold", "value"),
@@ -2317,12 +2379,13 @@ def populate_cross_filter_values(cross_field, model, min_val, threshold):
 def update_cluster_plot(model, min_val, threshold, field, plot_type, top_n, groups, cross_field, cross_value, era):
     empty_fig = go.Figure()
     empty_fig.update_layout(paper_bgcolor="#ffffff", plot_bgcolor="#f9fafc")
+    no_cards = html.Div()
     if not all([model, min_val, threshold, field, plot_type]):
-        return empty_fig, "Select all parameters above."
+        return empty_fig, "Select all parameters above.", no_cards
 
     cdf = _load_cluster_csv(model, min_val, threshold)
     if cdf is None:
-        return empty_fig, f"Data file not found for model={model} min={min_val} t={threshold}"
+        return empty_fig, f"Data file not found for model={model} min={min_val} t={threshold}", no_cards
 
     # Apply era filter based on PMID as publication year proxy
     if era:
@@ -2332,18 +2395,18 @@ def update_cluster_plot(model, min_val, threshold, field, plot_type, top_n, grou
             pmid_num = pd.to_numeric(cdf["key"], errors="coerce")
             cdf = cdf[(pmid_num >= lo) & (pmid_num < hi)]
         if cdf.empty:
-            return empty_fig, f"No data found for era '{era}'."
+            return empty_fig, f"No data found for era '{era}'.", no_cards
 
     # Apply cross-field filter: keep only (key, protein_index) in the selected cross-field cluster
     if cross_field and cross_value:
         keep = cdf[(cdf["field"] == cross_field) & (cdf["cluster_label"] == cross_value)][["key", "protein_index"]]
         if keep.empty:
-            return empty_fig, f"No proteins found for {cross_field} = '{cross_value}'."
+            return empty_fig, f"No proteins found for {cross_field} = '{cross_value}'.", no_cards
         cdf = cdf.merge(keep, on=["key", "protein_index"], how="inner")
 
     field_df = cdf[cdf["field"] == field].copy()
     if field_df.empty:
-        return empty_fig, f"No data for field '{field}' in this parameter combination."
+        return empty_fig, f"No data for field '{field}' in this parameter combination.", no_cards
 
     # Filter by group if selected
     if groups:
@@ -2354,7 +2417,9 @@ def update_cluster_plot(model, min_val, threshold, field, plot_type, top_n, grou
             )
         ]
         if field_df.empty:
-            return empty_fig, f"No data for the selected group(s) in this field."
+            return empty_fig, f"No data for the selected group(s) in this field.", no_cards
+
+    diversity = _compute_diversity_cards(field_df)
 
     n = top_n or 0
     suffix_parts = []
@@ -2366,9 +2431,9 @@ def update_cluster_plot(model, min_val, threshold, field, plot_type, top_n, grou
         suffix_parts.append(f"{cross_field.replace('_',' ')}={cross_value}")
     suffix = f" — {' | '.join(suffix_parts)}" if suffix_parts else ""
     if plot_type == "cluster":
-        return _umap_figure(field_df, field + suffix, top_n=n), ""
+        return _umap_figure(field_df, field + suffix, top_n=n), "", diversity
     else:
-        return _distribution_figure(field_df, field + suffix, top_n=n), ""
+        return _distribution_figure(field_df, field + suffix, top_n=n), "", diversity
 
 
 _CONDITION_COLS = [