- SMAC: Should I use Psi attack?
- Lan server on Minecraft not working Pls help
- How to connect kindles through mindcraft without using Xbox live?
- Show “featured” tab even when there are no questions with bounties
- Please help. I can't ask questions [duplicate]
- Inline suggested edit reviews are truncated in medium-width windows
- Sitecore deployment rollback
- Can I use abesse with hinc, inde, and others?
- How to transfer golem tokens from my zebpay wallet to my ethereum address
- Calling a payable contract function from truffle console
- Achilles invulnerability?
- Может ли союз быть общим элементом для частей ССП?
- «Срок годности: 2 года». Каким членом предложения является «2 года»?
- Organisation membership with Drupal user account
- Self-sustaining thermal systems
- How to use
to delete and go next region with yasnippet
- How would gravity affect a structure that completely encircled the earth
- Sending activation emails to all pending users
- log-based activation function
Word based perplexity from char-rnn model
I'm training a character based RNN model for text prediction and want to compare it to similar models. Since most literature uses word based perplexity as a performance metric, what would be the "proper" way to calculate word based perplexity from a character based model?