print on launch

app.py

@@ -7,7 +7,7 @@ import operator
 pd.options.plotting.backend = "plotly"
 
 
-TITLE = "Diffusion Faces Cluster Explorer"
+TITLE = "Diffusion Professions Cluster Explorer"
 
 clusters_dicts = dict(
     (num_cl, json.load(open(f"clusters/professions_to_clusters_{num_cl}.json")))
@@ -30,22 +30,27 @@ df_models = {
     "Dall-E 2": "DallE",
 }
 
+
 def describe_cluster(num_clusters, block="label"):
-    cl_dict= clusters_dicts[num_clusters]
+    cl_dict = clusters_dicts[num_clusters]
     labels_values = sorted(cl_dict.items(), key=operator.itemgetter(1))
     labels_values.reverse()
     total = float(sum(cl_dict.values()))
     lv_prcnt = list(
-        (item[0], round(item[1] * 100 / total, 0)) for item in labels_values)
+        (item[0], round(item[1] * 100 / total, 0)) for item in labels_values
+    )
     top_label = lv_prcnt[0][0]
-    description_string = "<span>The most represented %s is <b>%s</b>, making up about <b>%d%%</b> of the cluster.</span>" % (
-    to_string(block), to_string(top_label), lv_prcnt[0][1])
+    description_string = (
+        "<span>The most represented %s is <b>%s</b>, making up about <b>%d%%</b> of the cluster.</span>"
+        % (to_string(block), to_string(top_label), lv_prcnt[0][1])
+    )
     description_string += "<p>This is followed by: "
     for lv in lv_prcnt[1:]:
         description_string += "<BR/><b>%s:</b> %d%%" % (to_string(lv[0]), lv[1])
     description_string += "</p>"
     return description_string
 
+
 def make_profession_plot(num_clusters, prof_name):
     pre_pandas = dict(
         [
@@ -123,19 +128,24 @@ def make_profession_table(num_clusters, prof_names, mod_name, max_cols=8):
         for prof_name, prof_clusters in professions_list_clusters
     ]
     clusters_df = pd.DataFrame.from_dict(prof_list_pre_pandas)
-    return [c[0] for c in totals], (clusters_df.style.background_gradient(
-                axis=None, vmin=0, vmax=100, cmap="YlGnBu"
-            )
-            .format(precision=1)
-            .to_html()
+    return [c[0] for c in totals], (
+        clusters_df.style.background_gradient(
+            axis=None, vmin=0, vmax=100, cmap="YlGnBu"
         )
+        .format(precision=1)
+        .to_html()
+    )
+
 
 def show_examplars(num_clusters, prof_name, mod_name, cl_id):
     # TODO: show the actual images
-    examplars_dict = clusters_dicts[num_clusters][df_models[mod_name]][prof_name]["cluster_examplars"][str(cl_id)]
+    examplars_dict = clusters_dicts[num_clusters][df_models[mod_name]][prof_name][
+        "cluster_examplars"
+    ][str(cl_id)]
     return json.dumps(examplars_dict)
 
-with gr.Blocks() as demo:
+
+with gr.Blocks(title=TITLE) as demo:
     gr.Markdown("# 🤗 Diffusion Cluster Explorer")
     gr.Markdown("description will go here")
     with gr.Tab("Professions Overview"):
@@ -167,12 +177,18 @@ with gr.Blocks() as demo:
                     interactive=True,
                 )
             with gr.Column(scale=3):
-                with gr.Row():             
+                with gr.Row():
                     table = gr.HTML(
                         label="Profession assignment per cluster", wrap=True
-                    )   
-                    #clusters = gr.Dataframe(type="array", visible=False, col_count=1)
-                    clusters = gr.Textbox(label='clusters',visible=False)
+                    )
+                    # clusters = gr.Dataframe(type="array", visible=False, col_count=1)
+                    clusters = gr.Textbox(label="clusters", visible=False)
+                demo.load(
+                    make_profession_table,
+                    [num_clusters, profession_choices_overview, model_choices],
+                    [clusters, table],
+                    queue=False,
+                )
                 for var in [num_clusters, model_choices, profession_choices_overview]:
                     var.change(
                         make_profession_table,
@@ -181,8 +197,6 @@ with gr.Blocks() as demo:
                         queue=False,
                     )
 
-
-
     with gr.Tab("Profession Focus"):
         with gr.Row():
             with gr.Column():
@@ -197,7 +211,9 @@ with gr.Blocks() as demo:
                     value="social worker",
                     label="Select profession:",
                 )
-                gr.Markdown("You can show examples of profession images assigned to each cluster:")
+                gr.Markdown(
+                    "You can show examples of profession images assigned to each cluster:"
+                )
                 model_choices_focus = gr.Dropdown(
                     [
                         "All Models",
@@ -218,6 +234,12 @@ with gr.Blocks() as demo:
                 plot = gr.Plot(
                     label=f"Makeup of the cluster assignments for profession {profession_choice_focus}"
                 )
+                demo.load(
+                    make_profession_plot,
+                    [num_clusters_focus, profession_choice_focus],
+                    plot,
+                    queue=False,
+                )
                 for var in [num_clusters_focus, profession_choice_focus]:
                     var.change(
                         make_profession_plot,
@@ -226,14 +248,33 @@ with gr.Blocks() as demo:
                         queue=False,
                     )
         with gr.Row():
-            examplars_plot = gr.JSON()          # TODO: turn this into a plot with the actual images
+            examplars_plot = (
+                gr.JSON()
+            )  # TODO: turn this into a plot with the actual images
+            demo.load(
+                show_examplars,
+                [
+                    num_clusters_focus,
+                    profession_choice_focus,
+                    model_choices_focus,
+                    cluster_id_focus,
+                ],
+                examplars_plot,
+                queue=False,
+            )
             for var in [model_choices_focus, cluster_id_focus]:
                 var.change(
                     show_examplars,
-                    [num_clusters_focus, profession_choice_focus, model_choices_focus, cluster_id_focus],
+                    [
+                        num_clusters_focus,
+                        profession_choice_focus,
+                        model_choices_focus,
+                        cluster_id_focus,
+                    ],
                     examplars_plot,
                     queue=False,
                 )
 
 
-demo.launch()
+if __name__ == "__main__":
+    demo.queue().launch(debug=True)