change file structure

Makefile

@@ -2,23 +2,23 @@ export CC  = gcc
 export CXX = g++
 export MPICXX = mpicxx
 export LDFLAGS=
-export CFLAGS = -Wall -O3 -msse2  -Wno-unknown-pragmas -fPIC -I../src
+export CFLAGS = -Wall -O3 -msse2  -Wno-unknown-pragmas -fPIC -Iinclude
 
 BPATH=lib
 # objectives that makes up rabit library
 MPIOBJ= $(BPATH)/engine_mpi.o
 OBJ= $(BPATH)/allreduce_base.o $(BPATH)/allreduce_robust.o $(BPATH)/engine.o $(BPATH)/engine_empty.o
 ALIB= lib/librabit.a lib/librabit_mpi.a lib/librabit_empty.a
-
+HEADERS=src/*.h include/*.h include/rabit/*.h
 .PHONY: clean all
 
 all: $(ALIB)
 
-$(BPATH)/allreduce_base.o: src/allreduce_base.cc src/*.h
+$(BPATH)/allreduce_base.o: src/allreduce_base.cc $(HEADERS)
-$(BPATH)/engine.o: src/engine.cc src/*.h
+$(BPATH)/engine.o: src/engine.cc $(HEADERS)
-$(BPATH)/allreduce_robust.o: src/allreduce_robust.cc src/*.h
+$(BPATH)/allreduce_robust.o: src/allreduce_robust.cc $(HEADERS)
-$(BPATH)/engine_mpi.o: src/engine_mpi.cc src/*.h
+$(BPATH)/engine_mpi.o: src/engine_mpi.cc $(HEADERS)
-$(BPATH)/engine_empty.o: src/engine_empty.cc src/*.h
+$(BPATH)/engine_empty.o: src/engine_empty.cc $(HEADERS)
 
 lib/librabit.a: $(BPATH)/allreduce_base.o $(BPATH)/allreduce_robust.o $(BPATH)/engine.o
 lib/librabit_empty.a: $(BPATH)/engine_empty.o

include/README.md

@@ -0,0 +1,7 @@
+Library Header Files of Rabit
+====
+* This folder contains all the header needed to use rabit libary
+* To use it, add include to the search path of the compiler
+* User only need to know [rabit.h](rabit.h) and [rabit_serializable.h](rabit_serializable.h) to use the library
+* Folder [rabit](rabit) contains headers for internal engine and implementation of template
+* Not all .h files in the projects are contained in include, .h files that are internally used by library remains at [src](../src)

include/rabit.h

@@ -16,11 +16,11 @@
 #include <functional>
 #endif // C++11
 // contains definition of ISerializable
-#include "./serializable.h"
+#include "./rabit_serializable.h"
 // engine definition of rabit, defines internal implementation
-// to use rabit interface, there is no need to read engine.h rabit.h and serializable.h
+// to use rabit interface, there is no need to read engine.h
-// is suffice to use the interface
+// rabit.h and serializable.h are suffice to use the interface
-#include "./engine.h"
+#include "./rabit/engine.h"
 
 /*! \brief namespace of rabit */
 namespace rabit {
@@ -282,5 +282,5 @@ class SerializeReducer {
 };
 }  // namespace rabit
 // implementation of template functions
-#include "./rabit-inl.h"
+#include "./rabit/rabit-inl.h"
 #endif  // RABIT_ALLREDUCE_H

include/rabit/engine.h

@@ -5,7 +5,7 @@
  */
 #ifndef RABIT_ENGINE_H
 #define RABIT_ENGINE_H
-#include "./serializable.h"
+#include "../rabit_serializable.h"
 
 namespace MPI {
 /*! \brief MPI data type just to be compatible with MPI reduce function*/
@@ -222,3 +222,4 @@ class ReduceHandle {
 }  // namespace engine
 }  // namespace rabit
 #endif  // RABIT_ENGINE_H
+

include/rabit/io.h

@@ -5,7 +5,7 @@
 #include <cstring>
 #include <string>
 #include "./utils.h"
-#include "./serializable.h"
+#include "../rabit_serializable.h"
 /*!
  * \file io.h
  * \brief utilities that implements different serializable interface

include/rabit/rabit-inl.h

@@ -9,6 +9,7 @@
 // use engine for implementation
 #include "./io.h"
 #include "./utils.h"
+#include "../rabit.h"
 
 namespace rabit {
 namespace engine {

include/rabit/timer.h

@@ -10,6 +10,7 @@
 
 namespace rabit {
 namespace utils {
+// TODO not net cross platform, avoid to use this in most places
 /*!
  * \brief return time in seconds 
  */

include/rabit_serializable.h

@@ -1,8 +1,8 @@
-#ifndef RABIT_SERIALIZABLE_H
+#ifndef RABIT_RABIT_SERIALIZABLE_H
-#define RABIT_SERIALIZABLE_H
+#define RABIT_RABIT_SERIALIZABLE_H
 #include <vector>
 #include <string>
-#include "./utils.h"
+#include "./rabit/utils.h"
 /*!
  * \file serializable.h
  * \brief defines serializable interface of rabit

src/README.md

@@ -0,0 +1,6 @@
+Source Files of Rabit
+====
+* This folder contains the source files of rabit library
+* The library headers are in folder [include](../include)
+* The .h files in this folder are internal header files that are only used by rabit and will not be seen by users
+

src/allreduce_base.h

@@ -13,9 +13,9 @@
 
 #include <vector>
 #include <string>
-#include "./utils.h"
+#include <rabit/utils.h>
+#include <rabit/engine.h>
 #include "./socket.h"
-#include "./engine.h"
 
 namespace MPI {
 // MPI data type to be compatible with existing MPI interface

src/allreduce_robust.cc

@@ -9,10 +9,11 @@
 #define NOMINMAX
 #include <limits>
 #include <utility>
-#include "./io.h"
+#include <rabit/io.h>
-#include "./utils.h"
+#include <rabit/utils.h>
+#include <rabit/engine.h>
+#include <rabit/rabit-inl.h>
 #include "./allreduce_robust.h"
-#include "./rabit.h"
 
 namespace rabit {
 namespace engine {

src/allreduce_robust.h

@@ -10,7 +10,7 @@
 #ifndef RABIT_ALLREDUCE_ROBUST_H
 #define RABIT_ALLREDUCE_ROBUST_H
 #include <vector>
-#include "./engine.h"
+#include <rabit/engine.h>
 #include "./allreduce_base.h"
 
 namespace rabit {

src/engine.cc

@@ -9,7 +9,7 @@
 #define _CRT_SECURE_NO_DEPRECATE
 #define NOMINMAX
 
-#include "./engine.h"
+#include <rabit/engine.h>
 #include "./allreduce_base.h"
 #include "./allreduce_robust.h"
 

src/engine_empty.cc

@@ -9,7 +9,8 @@
 #define _CRT_SECURE_NO_DEPRECATE
 #define NOMINMAX
 
-#include "./engine.h"
+#include <rabit/engine.h>
+
 namespace rabit {
 namespace engine {
 /*! \brief EmptyEngine */

src/engine_mpi.cc

@@ -9,8 +9,8 @@
 #define _CRT_SECURE_NO_DEPRECATE
 #define NOMINMAX
 #include <cstdio>
-#include "./engine.h"
+#include <rabit/engine.h>
-#include "./utils.h"
+#include <rabit/utils.h>
 #include <mpi.h>
 
 namespace rabit {

src/socket.h

@@ -21,7 +21,7 @@
 #endif
 #include <string>
 #include <cstring>
-#include "./utils.h"
+#include <rabit/utils.h>
 
 #if defined(_WIN32)
 typedef int ssize_t;

test/Makefile

@@ -1,30 +1,22 @@
 export CC  = gcc
 export CXX = g++
 export MPICXX = mpicxx
-export LDFLAGS= -pthread -lm  -lrt
+export LDFLAGS= -pthread -lm  -lrt -L../lib
-export CFLAGS = -Wall -O3 -msse2  -Wno-unknown-pragmas -fPIC -I../src  -std=c++11
+export CFLAGS = -Wall -O3 -msse2  -Wno-unknown-pragmas -fPIC -I../include  -std=c++11
 
 # specify tensor path
 BIN = speed_test test_model_recover test_local_recover
-# objectives that makes up rabit library
-RABIT_OBJ = allreduce_base.o allreduce_robust.o engine.o
-MPIOBJ = engine_mpi.o
-
 OBJ = $(RABIT_OBJ) speed_test.o test_model_recover.o test_local_recover.o
 MPIBIN = speed_test.mpi
-.PHONY: clean all
+.PHONY: clean all lib
 
 all: $(BIN) $(MPIBIN)
-# the rabit library
+lib:
-allreduce_base.o: ../src/allreduce_base.cc ../src/*.h
+	cd ..;make;cd -
-engine.o: ../src/engine.cc ../src/*.h
-engine_mpi.o: ../src/engine_mpi.cc
-allreduce_robust.o: ../src/allreduce_robust.cc ../src/*.h
-
 # programs 
-speed_test.o: speed_test.cpp ../src/*.h
+speed_test.o: speed_test.cpp ../include/*.h lib
-test_model_recover.o: test_model_recover.cpp ../src/*.h
+test_model_recover.o: test_model_recover.cpp ../include/*.h lib
-test_local_recover.o: test_local_recover.cpp ../src/*.h
+test_local_recover.o: test_local_recover.cpp ../include/*.h lib
 
 # we can link against MPI version to get use MPI
 speed_test: speed_test.o  $(RABIT_OBJ)
@@ -33,16 +25,13 @@ test_model_recover: test_model_recover.o  $(RABIT_OBJ)
 test_local_recover: test_local_recover.o  $(RABIT_OBJ)
 
 $(BIN) : 
-	$(CXX) $(CFLAGS) -o $@ $(filter %.cpp %.o %.c %.cc, $^) $(LDFLAGS)
+	$(CXX) $(CFLAGS) -o $@ $(filter %.cpp %.o %.c %.cc, $^) $(LDFLAGS) -lrabit
 
 $(OBJ) : 
 	$(CXX) -c $(CFLAGS) -o $@ $(firstword $(filter %.cpp %.c %.cc, $^) )
 
 $(MPIBIN) : 
-	$(MPICXX) $(CFLAGS) -o $@ $(filter %.cpp %.o %.c %.cc, $^) $(LDFLAGS)
+	$(MPICXX) $(CFLAGS) -o $@ $(filter %.cpp %.o %.c %.cc, $^) $(LDFLAGS) -lrabit_mpi
-
-$(MPIOBJ) : 
-	$(MPICXX) -c $(CFLAGS) -o $@ $(firstword $(filter %.cpp %.c %.cc, $^) )
 
 clean:
 	$(RM) $(OBJ) $(BIN) $(MPIBIN) $(MPIOBJ) *~ ../src/*~

test/config.h

@@ -10,7 +10,7 @@
 #include <string>
 #include <istream>
 #include <fstream>
-#include "./utils.h"
+#include "./rabit/utils.h"
 
 namespace rabit {
 namespace utils {

test/speed_test.cpp

@@ -1,7 +1,7 @@
 // This program is used to test the speed of rabit API
 #include <rabit.h>
-#include <utils.h>
+#include <rabit/utils.h>
-#include <timer.h>
+#include <rabit/timer.h>
 #include <cstdio>
 #include <cstdlib>
 #include <cmath>

test/test_local_recover.cpp

@@ -1,7 +1,7 @@
 // this is a test case to test whether rabit can recover model when 
 // facing an exception
 #include <rabit.h>
-#include <utils.h>
+#include <rabit/utils.h>
 #include <cstdio>
 #include <cstdlib>
 #include <cmath>

test/test_model_recover.cpp

@@ -1,7 +1,7 @@
 // this is a test case to test whether rabit can recover model when 
 // facing an exception
 #include <rabit.h>
-#include <utils.h>
+#include <rabit/utils.h>
 #include <cstdio>
 #include <cstdlib>
 #include <cmath>

toolkit/Makefile

@@ -2,7 +2,7 @@ export CC  = gcc
 export CXX = g++
 export MPICXX = mpicxx
 export LDFLAGS= -pthread -lm -L../lib
-export CFLAGS = -Wall -O3 -msse2  -Wno-unknown-pragmas -fPIC -I../src -std=c++11
+export CFLAGS = -Wall -O3 -msse2  -Wno-unknown-pragmas -fPIC -I../include -std=c++11
 
 # specify tensor path
 BIN = kmeans

toolkit/kmeans.cpp

@@ -1,7 +1,7 @@
 // this is a test case to test whether rabit can recover model when 
 // facing an exception
 #include <rabit.h>
-#include <utils.h>
+#include <rabit/utils.h>
 #include "./toolkit_util.h"
 #include <time.h>
 
@@ -105,32 +105,33 @@ int main(int argc, char *argv[]) {
     model.InitModel(num_cluster, data.feat_dim);
     InitCentroids(data, &model.centroids);
     model.Normalize();
-    utils::LogPrintf("[%d] start at %s\n",
+    rabit::TrackerPrintf("[%d] start at %s\n",
-                     rabit::GetRank(), rabit::GetProcessorName().c_str());
+                         rabit::GetRank(), rabit::GetProcessorName().c_str());
   } else {
-    utils::LogPrintf("[%d] restart iter=%d\n", rabit::GetRank(), iter);    
+    rabit::TrackerPrintf("[%d] restart iter=%d\n", rabit::GetRank(), iter);    
   }
   const unsigned num_feat = data.feat_dim;
   // matrix to store the result
   Matrix temp;
-  for (int r = iter; r < max_iter; ++r) { 
+  for (int r = iter; r < max_iter; ++r) {    
     temp.Init(num_cluster, num_feat + 1, 0.0f);    
-    // call allreduce
+    auto lazy_get_centroid = [&]() {
-    rabit::Allreduce<op::Sum>(&temp.data[0], temp.data.size(), [&]() {
+      // lambda function used to calculate the data if necessary
-        // lambda function used to calculate the data if necessary
+      // this function may not be called when the result can be directly recovered
-        // this function may not be called when the result can be directly recovered
+      const size_t ndata = data.NumRow();
-        const size_t ndata = data.NumRow();
+      for (size_t i = 0; i < ndata; ++i) {
-        for (size_t i = 0; i < ndata; ++i) {
+        SparseMat::Vector v = data[i];
-          SparseMat::Vector v = data[i];
+        size_t k = GetCluster(model.centroids, v);
-          size_t k = GetCluster(model.centroids, v);
+        // temp[k] += v
-          // temp[k] += v
+        for (size_t j = 0; j < v.length; ++j) {
-          for (size_t j = 0; j < v.length; ++j) {
+          temp[k][v[j].findex] += v[j].fvalue;
-            temp[k][v[j].findex] += v[j].fvalue;
-          }
-          // use last column to record counts
-          temp[k][num_feat] += 1.0f;
         }
-      });
+        // use last column to record counts
+        temp[k][num_feat] += 1.0f;
+      }
+    };
+    // call allreduce
+    rabit::Allreduce<op::Sum>(&temp.data[0], temp.data.size(), lazy_get_centroid);
     // set number
     for (int k = 0; k < num_cluster; ++k) {
       float cnt = temp[k][num_feat];
@@ -146,7 +147,7 @@ int main(int argc, char *argv[]) {
   if (rabit::GetRank() == 0) {
     model.centroids.Print(argv[4]);
   }
-  utils::LogPrintf("[%d] Time taken: %f seconds\n", rabit::GetRank(), static_cast<float>(clock() - tStart) / CLOCKS_PER_SEC);
+  rabit::TrackerPrintf("[%d] Time taken: %f seconds\n", rabit::GetRank(), static_cast<float>(clock() - tStart) / CLOCKS_PER_SEC);
   rabit::Finalize();
   return 0;
 }