Removed the dlib::sparse_vector namespace. I put everything from this

namespace into the normal dlib:: namespace so that code which works
with both sparse and dense vectors is more cohesive.

dlib/test/svm_c_linear.cpp

@@ -135,30 +135,30 @@ namespace
 
 
         // Now test some of the sparse helper functions
-        DLIB_TEST(sparse_vector::max_index_plus_one(samples) == 2);
-        DLIB_TEST(sparse_vector::max_index_plus_one(samples[0]) == 2);
+        DLIB_TEST(max_index_plus_one(samples) == 2);
+        DLIB_TEST(max_index_plus_one(samples[0]) == 2);
 
         matrix<double,3,1> m;
         m = 1;
-        sparse_vector::add_to(m, samples[3]);
+        add_to(m, samples[3]);
         DLIB_TEST(m(0) == 1 + samples[3][0].second);
         DLIB_TEST(m(1) == 1 + samples[3][1].second);
         DLIB_TEST(m(2) == 1);
 
         m = 1;
-        sparse_vector::subtract_from(m, samples[3]);
+        subtract_from(m, samples[3]);
         DLIB_TEST(m(0) == 1 - samples[3][0].second);
         DLIB_TEST(m(1) == 1 - samples[3][1].second);
         DLIB_TEST(m(2) == 1);
 
         m = 1;
-        sparse_vector::add_to(m, samples[3], 2);
+        add_to(m, samples[3], 2);
         DLIB_TEST(m(0) == 1 + 2*samples[3][0].second);
         DLIB_TEST(m(1) == 1 + 2*samples[3][1].second);
         DLIB_TEST(m(2) == 1);
 
         m = 1;
-        sparse_vector::subtract_from(m, samples[3], 2);
+        subtract_from(m, samples[3], 2);
         DLIB_TEST(m(0) == 1 - 2*samples[3][0].second);
         DLIB_TEST(m(1) == 1 - 2*samples[3][1].second);
         DLIB_TEST(m(2) == 1);
@@ -227,7 +227,6 @@ namespace
                 sv[0] = 1;
                 sv[3] = 1;
 
-                using namespace sparse_vector;
 
                 DLIB_TEST(dot(sv,dv) == 5);
                 DLIB_TEST(dot(dv,sv) == 5);
@@ -249,7 +248,6 @@ namespace
                 sv[0] = 1;
                 sv[3] = 1;
 
-                using namespace sparse_vector;
 
                 assign(dv2, dv);
 

dlib/test/svm_struct.cpp

@@ -64,7 +64,7 @@ namespace
             feature_vector_type& psi
         ) const 
         {
-            sparse_vector::assign(psi, samples[idx]);
+            assign(psi, samples[idx]);
             // Add a constant -1 to account for the bias term.
             psi.push_back(std::make_pair(dims-1,static_cast<scalar_type>(-1)));
 
@@ -88,8 +88,6 @@ namespace
             // LOSS(idx,y) + F(x,y).  Note that y in this case is given by distinct_labels[i].
             for (unsigned long i = 0; i < distinct_labels.size(); ++i)
             {
-                using dlib::sparse_vector::dot;
-                using dlib::dot;
                 // Compute the F(x,y) part:
                 // perform: temp == dot(relevant part of current solution, samples[idx]) - current_bias
                 scalar_type temp = dot(rowm(current_solution, range(i*dims, (i+1)*dims-2)), samples[idx]) - current_solution((i+1)*dims-1);
@@ -106,7 +104,7 @@ namespace
                 }
             }
 
-            sparse_vector::assign(psi, samples[idx]);
+            assign(psi, samples[idx]);
             // add a constant -1 to account for the bias term
             psi.push_back(std::make_pair(dims-1,static_cast<scalar_type>(-1)));
 
@@ -221,7 +219,7 @@ namespace
 
             trained_function_type df;
 
-            const long dims = sparse_vector::max_index_plus_one(all_samples);
+            const long dims = max_index_plus_one(all_samples);
             df.labels  = select_all_distinct_labels(all_labels);
             df.weights = colm(reshape(weights, df.labels.size(), dims+1), range(0,dims-1));
             df.b       = colm(reshape(weights, df.labels.size(), dims+1), dims);
@@ -302,7 +300,7 @@ namespace
 
             trained_function_type df;
 
-            const long dims = sparse_vector::max_index_plus_one(all_samples);
+            const long dims = max_index_plus_one(all_samples);
             df.labels  = select_all_distinct_labels(all_labels);
             df.weights = colm(reshape(weights, df.labels.size(), dims+1), range(0,dims-1));
             df.b       = colm(reshape(weights, df.labels.size(), dims+1), dims);
@@ -383,7 +381,7 @@ namespace
 
             trained_function_type df;
 
-            const long dims = sparse_vector::max_index_plus_one(all_samples);
+            const long dims = max_index_plus_one(all_samples);
             df.labels  = select_all_distinct_labels(all_labels);
             df.weights = colm(reshape(weights, df.labels.size(), dims+1), range(0,dims-1));
             df.b       = colm(reshape(weights, df.labels.size(), dims+1), dims);
@@ -464,7 +462,7 @@ namespace
 
             trained_function_type df;
 
-            const long dims = sparse_vector::max_index_plus_one(all_samples);
+            const long dims = max_index_plus_one(all_samples);
             df.labels  = select_all_distinct_labels(all_labels);
             df.weights = colm(reshape(weights, df.labels.size(), dims+1), range(0,dims-1));
             df.b       = colm(reshape(weights, df.labels.size(), dims+1), dims);