Remove WMD

Author: jwijffels

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: doc2vec
 Type: Package
 Title: Distributed Representations of Sentences and Documents
-Version: 0.1.0
+Version: 0.1.1
 Authors@R: c(
     person('Jan', 'Wijffels', role = c('aut', 'cre', 'cph'), email = '[email protected]', comment = "R wrapper"), 
     person('BNOSAC', role = 'cph', comment = "R wrapper"),

NEWS.md

@@ -1,3 +1,9 @@
+## CHANGES IN doc2vec VERSION 0.1.1
+
+- Fixes for valgrind R CMD checks 
+    - Fixes for destructors of Vocabulary
+    - Remove WMD
+
 ## CHANGES IN doc2vec VERSION 0.1.0
 
 - Initial package based on https://github.com/hiyijian/doc2vec commit dec123e891f17ea664053ee7575b0e5e7dae4fca

src/Makevars

@@ -2,7 +2,7 @@ CXX_STD = CXX11
 PKG_LIBS = -pthread
 PKG_CPPFLAGS = -pthread -DSTRICT_R_HEADERS -I./doc2vec 
 
-SOURCES = doc2vec/Doc2Vec.cpp \
+#SOURCES = doc2vec/Doc2Vec.cpp \
 			doc2vec/NN.cpp \
 			doc2vec/TaggedBrownCorpus.cpp \
 			doc2vec/TrainModelThread.cpp \
@@ -11,6 +11,14 @@ SOURCES = doc2vec/Doc2Vec.cpp \
 			rcpp_doc2vec.cpp \
 			RcppExports.cpp
 
+SOURCES = doc2vec/Doc2Vec.cpp \
+			doc2vec/NN.cpp \
+			doc2vec/TaggedBrownCorpus.cpp \
+			doc2vec/TrainModelThread.cpp \
+			doc2vec/Vocab.cpp \
+			rcpp_doc2vec.cpp \
+			RcppExports.cpp			
+			
 OBJECTS = $(SOURCES:.cpp=.o)
 
 .PHONY: all 

src/doc2vec/Doc2Vec.cpp

@@ -3,7 +3,7 @@
 #include "Doc2Vec.h"
 #include "NN.h"
 #include "Vocab.h"
-#include "WMD.h"
+//#include "WMD.h"
 #include "TrainModelThread.h"
 #include "TaggedBrownCorpus.h"
 
@@ -17,7 +17,7 @@ void * trainModelThread(void * params)
 }
 
 /////==============================DOC2VEC========================
-Doc2Vec::Doc2Vec(): m_word_vocab(NULL), m_doc_vocab(NULL), m_nn(NULL), m_wmd(NULL),
+Doc2Vec::Doc2Vec(): m_word_vocab(NULL), m_doc_vocab(NULL), m_nn(NULL), //m_wmd(NULL),
   m_brown_corpus(NULL), m_expTable(NULL), m_negtive_sample_table(NULL)
 {
   initExpTable();
@@ -28,7 +28,7 @@ Doc2Vec::~Doc2Vec()
   if(m_word_vocab) delete m_word_vocab;
   if(m_doc_vocab) delete m_doc_vocab;
   if(m_nn) delete m_nn;
-  if(m_wmd) delete m_wmd;
+  //if(m_wmd) delete m_wmd;
   if(m_brown_corpus) delete m_brown_corpus;
   if(m_expTable) free(m_expTable);
   if(m_negtive_sample_table) free(m_negtive_sample_table);
@@ -99,11 +99,11 @@ void Doc2Vec::train(const char * train_file,
   free(pt);
   if(m_trace > 0){
     std::time_t t = std::chrono::system_clock::to_time_t(std::chrono::system_clock::now());
-    Rcpp::Rcout << Rcpp::as<Rcpp::Datetime>(Rcpp::wrap(t)) << " Closed all threads, normalising & WMD" << "\n";     
+    Rcpp::Rcout << Rcpp::as<Rcpp::Datetime>(Rcpp::wrap(t)) << " Closed all threads, normalising" << "\n";     
   }
   m_nn->norm();
-  m_wmd = new WMD(this);
-  m_wmd->train();
+  //m_wmd = new WMD(this);
+  //m_wmd->train();
 }
 
 void Doc2Vec::initTrainModelThreads(const char * train_file, int threads, int iter)
@@ -282,7 +282,7 @@ void Doc2Vec::save(FILE * fout)
   fwrite(&m_start_alpha, sizeof(real), 1, fout);
   fwrite(&m_sample, sizeof(real), 1, fout);
   fwrite(&m_iter, sizeof(int), 1, fout);
-  m_wmd->save(fout);
+  //m_wmd->save(fout);
 }
 
 void Doc2Vec::load(FILE * fin)
@@ -304,12 +304,12 @@ void Doc2Vec::load(FILE * fin)
   if(errnr <= 0) Rcpp::stop("fread failed");
   initNegTable();
   m_nn->norm();
-  m_wmd = new WMD(this);
-  m_wmd->load(fin);
+  //m_wmd = new WMD(this);
+  //m_wmd->load(fin);
 }
 
 long long Doc2Vec::dim() {return m_nn->m_dim;}
-WMD * Doc2Vec::wmd() {return m_wmd;}
+//WMD * Doc2Vec::wmd() {return m_wmd;}
 Vocabulary* Doc2Vec::wvocab() {return m_word_vocab;}
 Vocabulary* Doc2Vec::dvocab() {return m_doc_vocab;}
 NN * Doc2Vec::nn() {return m_nn;}

src/doc2vec/Doc2Vec.h

@@ -6,15 +6,15 @@
 class TrainModelThread;
 class NN;
 class Vocabulary;
-class WMD;
+//class WMD;
 class TaggedBrownCorpus;
 class TaggedDocument;
 struct knn_item_t;
 
 class Doc2Vec
 {
 friend class TrainModelThread;
-friend class WMD;
+//friend class WMD;
 friend class UnWeightedDocument;
 friend class WeightedDocument;
 public:
@@ -30,7 +30,7 @@ friend class WeightedDocument;
   Vocabulary* wvocab();
   Vocabulary* dvocab();
   NN * nn();
-  WMD * wmd();
+  //WMD * wmd();
 
 public:
   real doc_likelihood(TaggedDocument * doc, int skip = -1);
@@ -59,7 +59,7 @@ friend class WeightedDocument;
   Vocabulary * m_word_vocab;
   Vocabulary * m_doc_vocab;
   NN * m_nn;
-  WMD * m_wmd;
+  //WMD * m_wmd;
   int m_cbow;
   int m_hs;
   int m_negtive;