consolidated gd_mf predict functions

1 files changed, 23 insertions, 32 deletions

diff --git a/vowpalwabbit/gd_mf.cc b/vowpalwabbit/gd_mf.cc
index 5230b8e6..9a4b93a4 100644
--- a/vowpalwabbit/gd_mf.cc
+++ b/vowpalwabbit/gd_mf.cc
@@ -28,9 +28,9 @@ namespace GDMF {
     vw* all;
   };
 
-void mf_local_predict(example* ec, regressor& reg);
+void mf_print_audit_features(vw& all, example* ec, size_t offset);
 
-float mf_inline_predict(vw& all, example* &ec)
+float mf_predict(vw& all, example* ec)
 {
   float prediction = all.p->lp->get_initial(ec->ld);
 
@@ -79,13 +79,30 @@ float mf_inline_predict(vw& all, example* &ec)
     cerr << "cannot use triples in matrix factorization" << endl;
     throw exception();
   }
-    
+
   // ec->topic_predictions has linear, x_dot_l_1, x_dot_r_1, x_dot_l_2, x_dot_r_2, ... 
 
-  return prediction;
+  ec->partial_prediction = prediction;
+
+  // finalize prediction and compute loss
+  label_data* ld = (label_data*)ec->ld;
+  all.set_minmax(all.sd, ld->label);
+
+  ec->final_prediction = GD::finalize_prediction(all, ec->partial_prediction);
+
+  if (ld->label != FLT_MAX)
+    {
+      ec->loss = all.loss->getLoss(all.sd, ec->final_prediction, ld->label) * ld->weight;
+    }
+
+  if (all.audit)
+    mf_print_audit_features(all, ec, 0);
+
+  return ec->final_prediction;
 }
 
-void mf_inline_train(vw& all, example* &ec, float update)
+
+void mf_train(vw& all, example* &ec, float update)
 {
       weight* weights = all.reg.weight_vector;
       size_t mask = all.reg.weight_mask;
@@ -190,32 +207,6 @@ void mf_print_audit_features(vw& all, example* ec, size_t offset)
   mf_print_offset_features(all, ec, offset);
 }
 
-void mf_local_predict(vw& all, example* ec)
-{
-  label_data* ld = (label_data*)ec->ld;
-  all.set_minmax(all.sd, ld->label);
-
-  ec->final_prediction = GD::finalize_prediction(all, ec->partial_prediction);
-
-  if (ld->label != FLT_MAX)
-    {
-      ec->loss = all.loss->getLoss(all.sd, ec->final_prediction, ld->label) * ld->weight;
-    }
-
-  if (all.audit)
-    mf_print_audit_features(all, ec, 0);
-}
-
-float mf_predict(vw& all, example* ex)
-{
-  float prediction = mf_inline_predict(all, ex);
-
-  ex->partial_prediction = prediction;
-  mf_local_predict(all, ex);
-
-  return ex->final_prediction;
-}
-
   void save_load(void* d, io_buf& model_file, bool read, bool text)
 {
   vw* all = ((gdmf*)d)->all;
@@ -287,7 +278,7 @@ void end_pass(void* d)
  
     mf_predict(*all,ec);
     if (all->training && ((label_data*)(ec->ld))->label != FLT_MAX)
-      mf_inline_train(*all, ec, ec->eta_round);
+      mf_train(*all, ec, ec->eta_round);
   }
 
   // placeholder