Yes, this is how it is supposed to work. Even when not hashed, as numeric features are, they still must fit in the weight-vector, so there's always a moduloariel_vw
For some reason VW seems to hash all of my features not just the strings. Here is a simple example: echo '0.1726 | 2374907 5086457 1223443:5' | ./vw -aigor_tatarinov
Hi Ariel, Thanks for your reply. Here is the progress output that VW gives (after two passes): final_regressor = temp.model Num weight bits = 18 learning rate
- Open Source
- Feb 15, 2011
- This is a public group.
- Attachments are permitted.
- Members cannot hide email address.
- Listed in Yahoo Groups directory.
- Membership does not require approval.
- Messages from new members require approval.
- Anyone can post messages.