chore: apply security quarantine patch

Author: Cedric-Magnan

lib/pages/page.py

@@ -74,6 +74,7 @@ def write():
         column_name = column_select.selectbox(
             "columns", options=["-"] + dataset.columns.tolist()
         )
+        
         sample_data = column_select.checkbox("Sample a smaller dataset")
         remove_stopwords = column_select.checkbox("Remove stopwords")
         column_select.info(
@@ -145,16 +146,6 @@ def write():
         else:
             xi_value = 0.01
 
-        ##############################
-        #       VARIOUS BUTTONS      #
-        ##############################
-
-    space = streamlit.sidebar.text("")
-    compute = streamlit.sidebar.button("Compute embeddings")
-    show_labeled = streamlit.sidebar.checkbox("Show labeled data")
-    clear_cache_button = streamlit.sidebar.button("Clear cache")
-    clear_labels_button = streamlit.sidebar.button("Clear labels")
-    generate_sample_wordcloud = streamlit.sidebar.checkbox("Generate wordcloud")
 
     ##################################
     #      EXPORTING DATA            #
@@ -172,6 +163,20 @@ def write():
         "Exported data is saved by default to data/labeled_data/your_name.csv. modify this in the config file"
     )
 
+        ##############################
+        #       VARIOUS BUTTONS      #
+        ##############################
+
+    space = streamlit.sidebar.text("")
+    compute = streamlit.sidebar.button("Compute embeddings")
+    clear_cache_button = streamlit.sidebar.button("Clear cache")
+    clear_labels_button = streamlit.sidebar.button("Clear labels")
+
+    show_labeled = streamlit.sidebar.checkbox("Show labeled data")
+    
+    generate_sample_wordcloud = streamlit.sidebar.checkbox("Generate wordcloud")
+
+
     if clear_cache_button:
         clear_cache()
 
@@ -194,6 +199,8 @@ def write():
     #####################################
 
     if column_name != "-":
+
+        dataset[column_name]=dataset[column_name].astype(str)
         embedding_df = compute_cache(
             progress_bar,
             embedding_language,

lib/utils/cache.py

@@ -16,12 +16,14 @@
 def compute_stopwords(
     remove_stopwords, sampled_data, column_name, dataset_language, custom_stopwords
 ):
+    
     if remove_stopwords:
-        dataset_to_filter = sampled_data.copy()
-        
-        dataset_to_filter[column_name] = remove_stopwords_from_column(
-            dataset_to_filter[column_name], dataset_language, custom_stopwords
-        )
+        with streamlit.spinner(':hourglass: Filtering Stopwords...'):
+            dataset_to_filter = sampled_data.copy()
+            
+            dataset_to_filter[column_name] = remove_stopwords_from_column(
+                dataset_to_filter[column_name], dataset_language, custom_stopwords
+            )
         return dataset_to_filter
     else:
         return sampled_data
@@ -107,15 +109,12 @@ def compute_all(
 ):
     df = compute_samples(sample_data, df)
     # write_cache("sampled_cache.csv", df)
-    streamlit.write(df)
     df = compute_stopwords(remove_stopwords, df, column_name,dataset_language,custom_stopwords)
     # write_cache("stopwords_cache.csv", df)
-    streamlit.write(df)
     df = compute_embeddings(
         df, embedding_language, languages_dict, progress_bar, column_name
     )
     # write_cache("embedding_cache.csv", df)
-    streamlit.write(df)
     df = compute_dimension_reduction(df, transformer_option, transformers_dict)
     # write_cache("cache.csv",df)
     return df

lib/utils/plotting.py

@@ -49,7 +49,7 @@ def make_interactive_plot(embset, cluster):
     plot.xgrid.grid_line_color = None
     plot.ygrid.grid_line_color = None
     plot.outline_line_color = None
-
+    plot.toolbar.logo = None
 
 
 

lib/utils/processing.py

@@ -99,7 +99,7 @@ def cluster_tfidf(df, language="French"):
             colormap="Blues",
             width=100,
             height=100,
-        ).generate(" ".join(df[df.cluster == k].text.tolist()).lower())
+        ).generate(" ".join(df[df.cluster == k].text.astype(str).tolist()).lower())
         top_3 = list(wordcloud.words_.keys())
         df.loc[df.cluster == k, "cluster"] = f"{top_3[0]} - {top_3[1]} - {top_3[2]}"
     return df

main_app.py

@@ -13,3 +13,4 @@ def main():
 if __name__ == "__main__":
     st.set_page_config(layout='wide')
     main()
+