fix sometimes python cachelist problem

python/xgboost_python.cpp

@@ -112,7 +112,7 @@ namespace xgboost{
         private:
             bool init_trainer, init_model;
         public:
-            Booster(const std::vector<const regrank::DMatrix *> mats){
+            Booster(const std::vector<regrank::DMatrix *> mats){
                 silent = 1;
                 init_trainer = false;
                 init_model = false;
@@ -223,7 +223,7 @@ extern "C"{
 
     // xgboost implementation
     void *XGBoosterCreate( void *dmats[], size_t len ){
-        std::vector<const xgboost::regrank::DMatrix*> mats;
+        std::vector<xgboost::regrank::DMatrix*> mats;
         for( size_t i = 0; i < len; ++i ){
             DMatrix *dtr = static_cast<DMatrix*>(dmats[i]);
             dtr->CheckInit();

regrank/xgboost_regrank.h

@@ -31,7 +31,7 @@ namespace xgboost{
              * \brief a regression booter associated with training and evaluating data
              * \param mats  array of pointers to matrix whose prediction result need to be cached
              */
-            RegRankBoostLearner(const std::vector<const DMatrix *>& mats){
+            RegRankBoostLearner(const std::vector<DMatrix *>& mats){
                 silent = 0;
                 obj_ = NULL;
                 name_obj_ = "reg:linear";
@@ -45,7 +45,7 @@ namespace xgboost{
              *             data matrices to continue training otherwise it will cause error
              * \param mats  array of pointers to matrix whose prediction result need to be cached
              */          
-            inline void SetCacheData(const std::vector<const DMatrix *>& mats){
+            inline void SetCacheData(const std::vector<DMatrix *>& mats){
                 // estimate feature bound
                 int num_feature = 0;
                 // assign buffer index
@@ -58,7 +58,9 @@ namespace xgboost{
                         if( mats[i] == mats[j] ) dupilicate = true;
                     }
                     if( dupilicate ) continue;
-                    cache_.push_back( CacheEntry( mats[i], buffer_size ) );
+                    // set mats[i]'s cache learner pointer to this
+                    mats[i]->cache_learner_ptr_ = this;
+                    cache_.push_back( CacheEntry( mats[i], buffer_size, mats[i]->Size() ) );
                     buffer_size += static_cast<unsigned>(mats[i]->Size());
                     num_feature = std::max(num_feature, (int)(mats[i]->data.NumCol()));
                 }
@@ -342,9 +344,10 @@ namespace xgboost{
         private:
             struct CacheEntry{
                 const DMatrix *mat_;
-                int buffer_offset_;
-                CacheEntry(const DMatrix *mat, int buffer_offset)
-                    :mat_(mat), buffer_offset_(buffer_offset){}
+                int   buffer_offset_;
+                size_t num_row_; 
+                CacheEntry(const DMatrix *mat, int buffer_offset, size_t num_row)
+                    :mat_(mat), buffer_offset_(buffer_offset), num_row_(num_row){}
             };           
             /*! \brief the entries indicates that we have internal prediction cache */
             std::vector<CacheEntry> cache_;
@@ -352,7 +355,14 @@ namespace xgboost{
             // find internal bufer offset for certain matrix, if not exist, return -1
             inline int FindBufferOffset(const DMatrix &mat){
                 for(size_t i = 0; i < cache_.size(); ++i){
-                    if( cache_[i].mat_ == &mat ) return cache_[i].buffer_offset_; 
+                    if( cache_[i].mat_ == &mat && mat.cache_learner_ptr_ == this ) {
+                        if( cache_[i].num_row_ == mat.Size() ){                            
+                            return cache_[i].buffer_offset_; 
+                        }else{
+                            fprintf( stderr, "warning: number of rows in input matrix changed as remembered in cachelist, ignore cached results\n" );
+                            fflush( stderr );
+                        }
+                    }
                 }
                 return -1;
             } 

regrank/xgboost_regrank_data.h

@@ -52,9 +52,15 @@ namespace xgboost{
             booster::FMatrixS data;
             /*! \brief information fields */
             Info info;
+            /*! 
+             * \brief cache pointer to verify if the data structure is cached in some learner 
+             * this is a bit ugly, we need to have double check verification, so if one side get deleted, 
+             * and some strange re-allocation gets the same pointer we will still be fine
+             */
+            void *cache_learner_ptr_;
         public:
             /*! \brief default constructor */
-            DMatrix(void){}
+            DMatrix(void):cache_learner_ptr_(NULL){}
             /*! \brief get the number of instances */
             inline size_t Size() const{
                 return data.NumRow();

regrank/xgboost_regrank_main.cpp

@@ -126,7 +126,7 @@ namespace xgboost{
                         deval.back()->CacheLoad(eval_data_paths[i].c_str(), silent != 0, use_buffer != 0);
                         devalall.push_back(deval.back());
                     }
-                    std::vector<const DMatrix *> dcache(1, &data);
+                    std::vector<DMatrix *> dcache(1, &data);
                     for( size_t i = 0; i < deval.size(); ++ i){
                         dcache.push_back( deval[i] );
                     }