local/gae/train.py didn't predict the number of clusters of a person

[sevenights]

I have read your paper and the code. But I'm curious about the process since you did the clustering first then the prediction of the size of the clusters.

In local/gae/train.py, the arg n_clusters is from {}_pubs_content.txt in local/gae/input_data.py. So it is not from the cluster_size/count.py?

I have run the script, and the process cluster_size/count.py take too long. Could you tell me the precision of count.py? and how it affects the final clustering( clustering without knowing the n_cluster in advance)

Thanks!

[sevenights]

and I also notice that when you train the word2vec model, there is a code random.shuffle(author_feature). Since you ignore the order of words in the sentence. But the order of words has an effect in the word2vec model.

[sanlunainiu]

I have read your paper and the code. But I'm curious about the process since you did the clustering first then the prediction of the size of the clusters.

In local/gae/train.py, the arg n_clusters is from {}_pubs_content.txt in local/gae/input_data.py. So it is not from the cluster_size/count.py?

I have run the script, and the process cluster_size/count.py take too long. Could you tell me the precision of count.py? and how it affects the final clustering( clustering without knowing the n_cluster in advance)

Thanks!

"May I ask if your confusion has been resolved?"