From 69d7f71ae8cea28dc44a932d8856a1301bea623e Mon Sep 17 00:00:00 2001
From: tqchen <tianqi.tchen@gmail.com>
Date: Fri, 19 Dec 2014 02:12:53 -0800
Subject: [PATCH] change kmeans to using lambda

---
 toolkit/Makefile   |  7 ++-----
 toolkit/kmeans.cpp | 29 ++++++++++++++++-------------
 toolkit/kmeans.sh  |  8 --------
 3 files changed, 18 insertions(+), 26 deletions(-)
 delete mode 100755 toolkit/kmeans.sh

diff --git a/toolkit/Makefile b/toolkit/Makefile
index 4d96e6bab..429264ad3 100644
--- a/toolkit/Makefile
+++ b/toolkit/Makefile
@@ -2,15 +2,12 @@ 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
+export CFLAGS = -Wall -O3 -msse2  -Wno-unknown-pragmas -fPIC -I../src -std=c++11
 
 # specify tensor path
 BIN = kmeans
 # objectives that makes up rabit library
-RABIT_OBJ = allreduce_base.o allreduce_robust.o engine.o
-MPIOBJ = engine_mpi.o
-
-OBJ = $(RABIT_OBJ) kmeans.o
+OBJ = kmeans.o
 MPIBIN = kmeans.mpi
 .PHONY: clean all lib
 
diff --git a/toolkit/kmeans.cpp b/toolkit/kmeans.cpp
index e6dffd500..c08e50c23 100644
--- a/toolkit/kmeans.cpp
+++ b/toolkit/kmeans.cpp
@@ -114,20 +114,23 @@ int main(int argc, char *argv[]) {
   // matrix to store the result
   Matrix temp;
   for (int r = iter; r < max_iter; ++r) { 
-    temp.Init(num_cluster, num_feat + 1, 0.0f);
-    const size_t ndata = data.NumRow();
-    for (size_t i = 0; i < ndata; ++i) {
-      SparseMat::Vector v = data[i];
-      size_t k = GetCluster(model.centroids, v);
-      // temp[k] += v
-      for (size_t j = 0; j < v.length; ++j) {
-        temp[k][v[j].findex] += v[j].fvalue;
-      }
-      // use last column to record counts
-      temp[k][num_feat] += 1.0f;
-    }
+    temp.Init(num_cluster, num_feat + 1, 0.0f);    
     // call allreduce
-    rabit::Allreduce<op::Sum>(&temp.data[0], temp.data.size());
+    rabit::Allreduce<op::Sum>(&temp.data[0], temp.data.size(), [&]() {
+        // lambda function used to calculate the data if necessary
+        // this function may not be called when the result can be directly recovered
+        const size_t ndata = data.NumRow();
+        for (size_t i = 0; i < ndata; ++i) {
+          SparseMat::Vector v = data[i];
+          size_t k = GetCluster(model.centroids, v);
+          // temp[k] += v
+          for (size_t j = 0; j < v.length; ++j) {
+            temp[k][v[j].findex] += v[j].fvalue;
+          }
+          // use last column to record counts
+          temp[k][num_feat] += 1.0f;
+        }
+      });
     // set number
     for (int k = 0; k < num_cluster; ++k) {
       float cnt = temp[k][num_feat];
diff --git a/toolkit/kmeans.sh b/toolkit/kmeans.sh
deleted file mode 100755
index 53235a318..000000000
--- a/toolkit/kmeans.sh
+++ /dev/null
@@ -1,8 +0,0 @@
-#!/bin/bash
-if [ "$#" -lt 4 ];
-then
-    echo "Usage <nslave> <k> <d> <itr> <data_dir>"
-    exit -1
-fi
-
-../submit_job.py $1 kmeans "${@:2}"