reconcile

1 files changed, 11 insertions, 1 deletions

diff --git a/vowpalwabbit/lda_core.cc b/vowpalwabbit/lda_core.cc
index c70c6551..fcf79afc 100644
--- a/vowpalwabbit/lda_core.cc
+++ b/vowpalwabbit/lda_core.cc
@@ -496,7 +496,7 @@ v_array<float> old_gamma;
       for (size_t k =0; k<all.lda; k++)

 	new_gamma[k] = new_gamma[k]*v[k]+all.lda_alpha;

     }

-  while (average_diff(all, old_gamma.begin, new_gamma.begin) > 0.001);

+  while (average_diff(all, old_gamma.begin, new_gamma.begin) > all.lda_epsilon);

 

   ec->topic_predictions.erase();

   ec->topic_predictions.resize(all.lda);

@@ -580,6 +580,15 @@ void save_load(lda& l, io_buf& model_file, bool read, bool text)
 

   void learn_batch(lda& l)

   {

+    if (l.sorted_features.empty()) {

+      // This can happen when the socket connection is dropped by the client.

+      // If l.sorted_features is empty, then l.sorted_features[0] does not

+      // exist, so we should not try to take its address in the beginning of

+      // the for loops down there. Since it seems that there's not much to

+      // do in this case, we just return.

+      return;

+    }

+

     float eta = -1;

     float minuseta = -1;

 

@@ -754,6 +763,7 @@ learner* setup(vw&all, vector<string>&opts, po::variables_map& vm)
     ("lda_alpha", po::value<float>(&all.lda_alpha), "Prior on sparsity of per-document topic weights")

     ("lda_rho", po::value<float>(&all.lda_rho), "Prior on sparsity of topic distributions")

     ("lda_D", po::value<float>(&all.lda_D), "Number of documents")

+    ("lda_epsilon", po::value<float>(&all.lda_epsilon), "Loop convergence threshold")

     ("minibatch", po::value<size_t>(&all.minibatch), "Minibatch size, for LDA");

 

   po::parsed_options parsed = po::command_line_parser(opts).