You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Jan 26, 2021. It is now read-only.
The current released lightlda doesn't support asymmetric Dirichlet prior optimization. However, our internal practice show it would be useful to get better model with such feature (Also see this).
If anyone is interested in contributing this feature, please reply or contact us through email. We can collaborate on this.
The text was updated successfully, but these errors were encountered:
Hi, guys. Thank you for your amazing work on large scale LDA.
On the other hand, I think model quality is as important as scalability. So I am very intresting in improving it. It is exciting to know asymmetric Dirichlet prior could help. Would you please to share some experience on this? I will try my best to contribute
Hi, guys,
I finished to try to add this new feature in PR#22
This PR supports asymmetric alpha in following steps:
Add two extra tables to Multiverso. One is topic frequency table, a matrix to count each topics’ frequency. The other one is doc length table, a row to count how many document is with length k.
Initialize the two extra tables with random initialized documents
Learn alpha distribution with the two extra table every 5 iterations
Build alias table for leanred alpha distribution
Sample topics with learned alpha distribution and alias table. Meanwhile, update countings of topic frequency table if necessary
To use this new feature, please just run with an extra option "-num_alpha_iterations".
Please notice that there are two TODOs. One is Evaluation in asymmetric prior mode, the other is Inference with asymmetric prior.
The current released lightlda doesn't support asymmetric Dirichlet prior optimization. However, our internal practice show it would be useful to get better model with such feature (Also see this).
If anyone is interested in contributing this feature, please reply or contact us through email. We can collaborate on this.
The text was updated successfully, but these errors were encountered: