Merge branch 'valueclass'

Author: kosloot

.github/workflows/timbl.yml

@@ -2,7 +2,7 @@ name: C/C++ CI
 
 on:
   push:
-    branches: [master]
+    branches: [master, valueclass]
     paths:
       - 'src/**'
       - 'include/**'
@@ -32,6 +32,7 @@ jobs:
             ${{ github.actor }} started a build of
             ${{ github.event.repository.name }}
             [${{ steps.extract_branch.outputs.branch }}]
+
   build:
     runs-on: ${{ matrix.os }}
     needs: notification

configure.ac

@@ -1,12 +1,12 @@
 #                                               -*- Autoconf -*-
 # Process this file with autoconf to produce a configure script.
 
-AC_PREREQ(2.61)
+AC_PREREQ([2.69])
 AC_INIT([timbl],[6.8],[[email protected]]) #also adapt in codemeta.json!
 AM_INIT_AUTOMAKE
 AC_CONFIG_SRCDIR([.])
 AC_CONFIG_MACRO_DIR([m4])
-AC_CONFIG_HEADER([config.h])
+AC_CONFIG_HEADERS([config.h])
 
 if test x"${CXXFLAGS+set}" = xset; then
   # the user set CXXFLAGS; don't override it.

demos/api_test6.cxx

@@ -34,12 +34,12 @@ using namespace Timbl;
 int main(){
   TimblAPI My_Experiment( "-a IB1 +vDI+DB -G 0 -k3", "test6" );
   My_Experiment.Learn( "dimin.train" );
-  const ValueDistribution *vd;
+  const ClassDistribution *vd;
   const TargetValue *tv
     = My_Experiment.Classify( std::string("-,=,O,m,+,h,K,=,-,n,I,N,K"), vd );
   cout << "resulting target: " << tv << endl;
   cout << "resulting Distribution: " << vd << endl;
-  ValueDistribution::dist_iterator it=vd->begin();
+  ClassDistribution::dist_iterator it=vd->begin();
   while ( it != vd->end() ){
     cout << it->second << " OR ";
     cout << it->second->Value() << " " << it->second->Weight() << endl;
@@ -48,7 +48,7 @@ int main(){
 
   cout << "the same with neighborSets" << endl;
   const neighborSet *nb = My_Experiment.classifyNS( "-,=,O,m,+,h,K,=,-,n,I,N,K" );
-  WValueDistribution *vd2 = nb->bestDistribution();
+  WClassDistribution *vd2 = nb->bestDistribution();
   vd2->Normalize();
   cout << "default answer " << vd2 << endl;
   decayStruct *dc = new  expDecay(0.3);

include/timbl/BestArray.h

@@ -38,15 +38,15 @@ namespace Timbl {
     friend std::ostream& operator<< ( std::ostream&, const BestRec * );
   public:
     BestRec();
+    BestRec( const BestRec& ) = delete; // forbid copies
+    BestRec& operator=( const BestRec& ) = delete; // forbid copies
     ~BestRec();
     size_t totalBests() const { return aggregateDist.totalSize(); };
     double bestDistance;
-    ValueDistribution aggregateDist;
-    std::vector<ValueDistribution*> bestDistributions;
+    ClassDistribution aggregateDist;
+    std::vector<ClassDistribution*> bestDistributions;
     std::vector<icu::UnicodeString> bestInstances;
   private:
-    BestRec( const BestRec& );
-    BestRec& operator=( const BestRec& );
   };
 
   class BestArray {
@@ -61,7 +61,7 @@ namespace Timbl {
     ~BestArray();
     void init( unsigned int, unsigned int, bool, bool, bool );
     double addResult( double,
-		      const ValueDistribution *,
+		      const ClassDistribution *,
 		      const icu::UnicodeString& );
     void initNeighborSet( neighborSet& ) const;
     void addToNeighborSet( neighborSet& , size_t ) const;

include/timbl/Choppers.h

@@ -34,8 +34,13 @@ namespace Timbl{
 
   class Chopper {
   public:
+    Chopper():
+      vSize(0)
+    {};
     virtual ~Chopper() {};
     virtual bool chop( const icu::UnicodeString&, size_t ) = 0;
+    const icu::UnicodeString& operator[]( int i ) const {
+      return choppedInput[i]; }
     const icu::UnicodeString& getField( size_t i ) const {
       return choppedInput[i];
     };
@@ -68,24 +73,32 @@ namespace Timbl{
 
   class ExChopper: public virtual Chopper {
   public:
-    double getExW() const { return exW; };
+    ExChopper():
+      Chopper(),
+      exW(-1.0)
+    {};
+    double getExW() const override { return exW; };
   protected:
-    void init( const icu::UnicodeString&, size_t, bool );
+    void init( const icu::UnicodeString&, size_t, bool ) override;
     double exW;
   };
 
   class OccChopper: public virtual Chopper {
   public:
-    int getOcc() const { return occ; };
+    OccChopper():
+      Chopper(),
+      occ(-1)
+    {};
+    int getOcc() const override { return occ; };
   protected:
-    void init( const icu::UnicodeString&, size_t, bool );
+    void init( const icu::UnicodeString&, size_t, bool ) override;
     int occ;
   };
 
   class C45_Chopper : public virtual Chopper {
   public:
-    bool chop( const icu::UnicodeString&, size_t );
-    icu::UnicodeString getString() const;
+    bool chop( const icu::UnicodeString&, size_t ) override;
+    icu::UnicodeString getString() const override;
   };
 
   class C45_ExChopper : public C45_Chopper, public ExChopper {
@@ -96,7 +109,7 @@ namespace Timbl{
 
   class ARFF_Chopper : public C45_Chopper {
   public:
-    bool chop( const icu::UnicodeString&, size_t );
+    bool chop( const icu::UnicodeString&, size_t ) override;
   };
 
   class ARFF_ExChopper : public C45_ExChopper {
@@ -107,8 +120,8 @@ namespace Timbl{
 
   class Bin_Chopper : public virtual Chopper {
   public:
-    bool chop( const icu::UnicodeString&, size_t );
-    icu::UnicodeString getString() const;
+    bool chop( const icu::UnicodeString&, size_t ) override;
+    icu::UnicodeString getString() const override;
   };
 
   class Bin_ExChopper : public Bin_Chopper, public ExChopper {
@@ -120,8 +133,8 @@ namespace Timbl{
   class Compact_Chopper : public virtual Chopper {
   public:
     explicit Compact_Chopper( int L ): fLen(L){};
-    bool chop( const icu::UnicodeString&, size_t );
-    icu::UnicodeString getString() const;
+    bool chop( const icu::UnicodeString&, size_t ) override;
+    icu::UnicodeString getString() const override;
   private:
     int fLen;
     Compact_Chopper();
@@ -143,8 +156,8 @@ namespace Timbl{
 
   class Columns_Chopper : public virtual Chopper {
   public:
-    bool chop( const icu::UnicodeString&, size_t );
-    icu::UnicodeString getString() const;
+    bool chop( const icu::UnicodeString&, size_t ) override;
+    icu::UnicodeString getString() const override;
   };
 
   class Columns_ExChopper : public Columns_Chopper, public ExChopper {
@@ -155,8 +168,8 @@ namespace Timbl{
 
   class Tabbed_Chopper : public virtual Chopper {
   public:
-    bool chop( const icu::UnicodeString&, size_t );
-    icu::UnicodeString getString() const;
+    bool chop( const icu::UnicodeString&, size_t ) override;
+    icu::UnicodeString getString() const override;
   };
 
   class Tabbed_ExChopper : public Tabbed_Chopper, public ExChopper {
@@ -168,8 +181,8 @@ namespace Timbl{
 
   class Sparse_Chopper : public virtual Chopper {
   public:
-    bool chop( const icu::UnicodeString&, size_t );
-    icu::UnicodeString getString() const;
+    bool chop( const icu::UnicodeString&, size_t ) override;
+    icu::UnicodeString getString() const override;
   };
 
   class Sparse_ExChopper : public Sparse_Chopper, public ExChopper {

include/timbl/Common.h

@@ -29,12 +29,13 @@
 #define TIMBL_COMMON_H
 
 #include <iostream>
+#include <limits>
 #include <cmath>
-#include <cfloat>
 #include <cassert>
 
 namespace Common {
-  const double Epsilon = DBL_EPSILON;   // smallest x so that 1+x != 1
+  const double Epsilon = std::numeric_limits<double>::epsilon();
+  // smallest x so that 1+x != 1
   const int DEFAULT_MAX_FEATS = 2500;   // default maximun number of Features
 
   std::string Version();