hpandeycodeit opened a new pull request #432: MADLIB-1351 : Added stopping 
criteria on perplexity to LDA
URL: https://github.com/apache/madlib/pull/432
 
 
   LDA:
   Added stopping criteria on perplexity to LDA. 
   
   [MADLIB-1351](MADLIB-1351)
   
   Currently, in LDA there are no stopping criteria. It runs for all the 
provided iterations. 
   This PR calculated the Perplexity on each iteration and when the difference 
between the last two Perplexity values is less than the perplexity_tol, it 
stops the iteration. 
   
   These are the two new parameters added to the function: 
   
   ```
   evaluate_every      Integer,
   perplexity_tol      Double Precision
   ```
   
   And there is a change to the Model output table as well. It will have these 
two extra columns 
   
   ```
   perplexity  DOUBLE PRECISION[]
   perplexity_iters INTEGER[]
   ```
   
   Where 
   perplexity is an Array of perplexity values as per the 'evaluate_every' 
parameter.
   perplexity_iters is an Array indicating the iterations for which perplexity 
is calculated
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

Reply via email to