diff --git a/topic generation/print_topics_per_cluster.py b/topic generation/print_topics_per_cluster.py new file mode 100644 index 0000000000000000000000000000000000000000..3b248c965d22873137ceeecd2213a8036806e06f --- /dev/null +++ b/topic generation/print_topics_per_cluster.py @@ -0,0 +1,10 @@ +def print_topics_per_cluster(df, agg_labels): + # Ajouter agg_labels comme colonne 'cluster' dans df + df['cluster'] = agg_labels + + print("\nTopics par cluster pour l'ensemble des données :") + for cluster in sorted(df['cluster'].unique()): + print(f"\nCluster {cluster}:") + messages_cluster = df[df['cluster'] == cluster]['message'].tolist() + topics = generate_topics_for_cluster(messages_cluster) + print("Topics générés:", topics) \ No newline at end of file