&#
- | Modifying Data 🚜 |
+ Altering Data ✂ |
@@ -470,6 +470,10 @@ API Reference with code samples &#
| modify_by_idx() |
+
+ | remove_bottom_n_data() |
+
+
| remove_column( 2 ) |
diff --git a/docs/HTML/remove_top_n_data.html b/docs/HTML/remove_top_n_data.html
index 407f419b..30ddebae 100644
--- a/docs/HTML/remove_top_n_data.html
+++ b/docs/HTML/remove_top_n_data.html
@@ -38,8 +38,8 @@
| Signature | Description | Parameters |
-
+
@@ -60,6 +60,26 @@
|
+
+
+
+template<comparable T, typename ... Ts>
+void
+remove_bottom_n_data(const char *col_name, size_type n);
+
+ |
+
+ It removes data rows corresponding to n bottom rows of the named column.
+ NOTE Comparison operators (<, >, ==) must be well defined for type T.
+ |
+
+ T: Type of the named column
+ Ts: The list of types for all columns. A type should be specified only once
+ col_name: Name of the data column
+ n: Number of bottom rows
+ |
+
+
static void test_remove_top_n_data() {
@@ -100,6 +120,47 @@
assert(df.get_column<int>("col_4")[2] == 24);
assert(view.get_column<int>("col_4")[2] == 24);
}
+
+// -----------------------------------------------------------------------------
+
+static void test_remove_bottom_n_data() {
+
+ std::cout << "\nTesting remove_bottom_n_data( ) ..." << std::endl;
+
+ StlVecType<unsigned long> idx = { 123450, 123451, 123452, 123453, 123454, 123455, 123456, 123457, 123458, 123459, 123460, 123461, 123462, 123463 };
+ StlVecType<double> d1 = { 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 };
+ StlVecType<double> d2 = { 8, 9, 10, 11, 12, 13, 14, 20, 22, 23, 30, 31, 32, 1.89 };
+ StlVecType<double> d3 = { 15, 16, 15, 18, 19, 16, 21, 0.34, 1.56, 0.34, 2.3, 0.34, 19.0, 10 };
+ StlVecType<int> i1 = { 22, 23, 24, 25, 99 };
+ MyDataFrame df;
+
+ df.load_data(std::move(idx),
+ std::make_pair("col_1", d1),
+ std::make_pair("col_2", d2),
+ std::make_pair("col_3", d3),
+ std::make_pair("col_4", i1));
+
+ MyDataFrame df2 = df;
+
+ auto lbd = [](const unsigned long &, const double &val) -> bool { return (val < 100.0); };
+ auto view = df2.get_view_by_sel<double, decltype(lbd), double, int, std::string>("col_1", lbd);
+
+ df.remove_bottom_n_data<double, int, double, std::string>("col_3", 4);
+ view.remove_bottom_n_data<double, int, double, std::string>("col_3", 4);
+
+ assert(df.get_index().size() == 10);
+ assert(view.get_index().size() == 10);
+ assert(df.get_column<double>("col_2").size() == 10);
+ assert(view.get_column<double>("col_2").size() == 10);
+ assert(df.get_column<int>("col_4").size() == 10);
+ assert(view.get_column<int>("col_4").size() == 10);
+ assert(df.get_index()[4] == 123454);
+ assert(view.get_index()[4] == 123454);
+ assert(df.get_column<double>("col_1")[6] == 7);
+ assert(view.get_column<double>("col_1")[6] == 7);
+ assert(df.get_column<int>("col_4")[2] == 24);
+ assert(view.get_column<int>("col_4")[2] == 24);
+}
, ==) must be well defined for type T.
+ //
+ // T:
+ // Type of column name
+ // Ts:
+ // List all the types of all data columns. A type should be specified in
+ // the list only once.
+ // col_name:
+ // Name of the given column
+ // n:
+ // Number of bottom rows
+ //
+ template
+ void
+ remove_bottom_n_data(const char *col_name, size_type n);
+
// It removes duplicate rows and returns a new DataFrame. Duplication is
// determined by the given column. remove_dup_spec determines which
// of the duplicated rows to keep.
diff --git a/include/DataFrame/Internals/DataFrame_remove.tcc b/include/DataFrame/Internals/DataFrame_remove.tcc
index 27110b39..69de9995 100644
--- a/include/DataFrame/Internals/DataFrame_remove.tcc
+++ b/include/DataFrame/Internals/DataFrame_remove.tcc
@@ -376,6 +376,36 @@ void DataFrame::remove_top_n_data(const char *col_name, size_type n) {
// ----------------------------------------------------------------------------
+template
+template
+void DataFrame::
+remove_bottom_n_data(const char *col_name, size_type n) {
+
+ static_assert(std::is_base_of, H>::value ||
+ std::is_base_of, H>::value,
+ "Only a StdDataFrame or a PtrView can call "
+ "remove_bottom_n_data()");
+
+ const ColumnVecType &vec = get_column(col_name);
+ NSmallestVisitor nsv { n };
+
+ nsv.pre();
+ nsv(indices_.begin(), indices_.end(), vec.begin(), vec.end());
+ nsv.post();
+ nsv.sort_by_index_idx();
+
+ StlVecType col_indices;
+
+ col_indices.reserve(n);
+ for (const auto &res : nsv.get_result())
+ col_indices.push_back(res.index_idx);
+
+ remove_data_by_sel_common_(col_indices);
+ return;
+}
+
+// ----------------------------------------------------------------------------
+
template
template
DataFrame> DataFrame::
diff --git a/test/dataframe_tester_4.cc b/test/dataframe_tester_4.cc
index a5bb27d8..213fa7da 100644
--- a/test/dataframe_tester_4.cc
+++ b/test/dataframe_tester_4.cc
@@ -1224,6 +1224,57 @@ static void test_remove_top_n_data() {
// -----------------------------------------------------------------------------
+static void test_remove_bottom_n_data() {
+
+ std::cout << "\nTesting remove_bottom_n_data( ) ..." << std::endl;
+
+ StlVecType idx =
+ { 123450, 123451, 123452, 123453, 123454, 123455, 123456,
+ 123457, 123458, 123459, 123460, 123461, 123462, 123463 };
+ StlVecType d1 =
+ { 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 };
+ StlVecType d2 =
+ { 8, 9, 10, 11, 12, 13, 14, 20, 22, 23, 30, 31, 32, 1.89 };
+ StlVecType d3 =
+ { 15, 16, 15, 18, 19, 16, 21, 0.34, 1.56, 0.34, 2.3, 0.34, 19.0, 10 };
+ StlVecType i1 = { 22, 23, 24, 25, 99 };
+ MyDataFrame df;
+
+ df.load_data(std::move(idx),
+ std::make_pair("col_1", d1),
+ std::make_pair("col_2", d2),
+ std::make_pair("col_3", d3),
+ std::make_pair("col_4", i1));
+
+ MyDataFrame df2 = df;
+
+ auto lbd =
+ [](const unsigned long &, const double &val) -> bool {
+ return (val < 100.0);
+ };
+ auto view =
+ df2.get_view_by_sel
+ ("col_1", lbd);
+
+ df.remove_bottom_n_data("col_3", 4);
+ view.remove_bottom_n_data("col_3", 4);
+
+ assert(df.get_index().size() == 10);
+ assert(view.get_index().size() == 10);
+ assert(df.get_column("col_2").size() == 10);
+ assert(view.get_column("col_2").size() == 10);
+ assert(df.get_column("col_4").size() == 10);
+ assert(view.get_column("col_4").size() == 10);
+ assert(df.get_index()[4] == 123454);
+ assert(view.get_index()[4] == 123454);
+ assert(df.get_column("col_1")[6] == 7);
+ assert(view.get_column("col_1")[6] == 7);
+ assert(df.get_column("col_4")[2] == 24);
+ assert(view.get_column("col_4")[2] == 24);
+}
+
+// -----------------------------------------------------------------------------
+
int main(int, char *[]) {
test_starts_with();
@@ -1245,6 +1296,7 @@ int main(int, char *[]) {
test_get_data_on_days_in_month();
test_get_data_between_times();
test_remove_top_n_data();
+ test_remove_bottom_n_data();
return (0);
}