Include Check to Hide All Smples Different to Real Ones

app.py

@@ -887,32 +887,36 @@ def run_model(model_name):
                         legend_label=f"Real: {label}",
                         source=source)
 
-        # Ploteamos los puntos de synthetic, diferenciando cada source con su marcador
-        if unique_subsets["synthetic"]:
-            for source_name, labels in unique_subsets["synthetic"].items():
-                df_source = df_all["synthetic"][df_all["synthetic"]["source"] == source_name]
-                marker = marker_mapping.get(source_name, "square")
-                # Para cada label en ese source, usamos la función auxiliar
-                renderers = add_synthetic_dataset_to_fig(fig_all, df_source, labels,
-                                                        marker=marker,
-                                                        color_mapping=color_maps["synthetic"][source_name],
-                                                        group_label=source_name)
-        # Ploteamos los puntos de pretrained (agrupados por label)
-        if unique_subsets["pretrained"]:
-            for label in unique_subsets["pretrained"]:
-                subset = df_all["pretrained"][df_all["pretrained"]['label'] == label]
-                source = ColumnDataSource(data={
-                    'x': subset['x'],
-                    'y': subset['y'],
-                    'label': subset['label'],
-                    'img': subset['img']
-                })
-                # Usamos 'triangle' para pretrained (por ejemplo)
-                fig_all.triangle('x', 'y', size=10,
-                                fill_color=color_maps["pretrained"][label],
-                                line_color=color_maps["pretrained"][label],
-                                legend_label=f"Pretrained: {label}",
-                                source=source)
+        show_real_only = st.checkbox("Show only real samples", value=False, key=f"show_real_only_{model_name}")
+
+        if not show_real_only:
+
+            # Ploteamos los puntos de synthetic, diferenciando cada source con su marcador
+            if unique_subsets["synthetic"]:
+                for source_name, labels in unique_subsets["synthetic"].items():
+                    df_source = df_all["synthetic"][df_all["synthetic"]["source"] == source_name]
+                    marker = marker_mapping.get(source_name, "square")
+                    # Para cada label en ese source, usamos la función auxiliar
+                    renderers = add_synthetic_dataset_to_fig(fig_all, df_source, labels,
+                                                            marker=marker,
+                                                            color_mapping=color_maps["synthetic"][source_name],
+                                                            group_label=source_name)
+            # Ploteamos los puntos de pretrained (agrupados por label)
+            if unique_subsets["pretrained"]:
+                for label in unique_subsets["pretrained"]:
+                    subset = df_all["pretrained"][df_all["pretrained"]['label'] == label]
+                    source = ColumnDataSource(data={
+                        'x': subset['x'],
+                        'y': subset['y'],
+                        'label': subset['label'],
+                        'img': subset['img']
+                    })
+                    # Usamos 'triangle' para pretrained (por ejemplo)
+                    fig_all.triangle('x', 'y', size=10,
+                                    fill_color=color_maps["pretrained"][label],
+                                    line_color=color_maps["pretrained"][label],
+                                    legend_label=f"Pretrained: {label}",
+                                    source=source)
 
         # Calcular el centroide y el radio (usando solo las muestras reales)
         center_x = df_all["real"]['x'].mean()