681830_zh-CN

liliconcom

脚本来选择包含值列表的所有列？

我正在挖掘一个大的 dt（70k cols，字符）。 大多数列都有广泛的值，但其中相当一部分只有两个令人讨厌的值。 我想识别所有这些列然后转储它们。

有什么建议吗？

 

这些令人讨厌的列将仅包含术语“**skipped**”或“dne”。

我已经转储了包含单个记录值的所有列。

 

 

自动化和脚本编写回复：用于选择包含值列表的所有列的脚本？

这是一个示例脚本

Names Default To Here( 1 );

// Create a sample data table to be used for illustration 
dt = New Table( "Example", add rows( 20 ) );
dt << delete columns( :Column 1 );
For( i = 1, i <= 100, i++,
 If( randomuniform( 0, 1 ) > .95,
  If( Random Uniform( 0, 1 ) > .5,
   dt << New Column( "col" || Char( i ),
    character,
    values( {"**skipped**"} )
   ),
   dt << New Column( "col" || Char( i ),
    character,
    values( {"dne"} )
   )
  ),
  dt << New Column( "col" || Char( i ),
   character,
   set each value( Char( Random Uniform( 0, 100 ) ) )
  )
 )
);

// Below is the simple script that will build a list of columns to be deleted,
// and then delete them
colNames = dt << get column names(string,character);

colsToDelete = {};
For Each( {name}, colnames,
 If(
  N Rows(
   (dt << get rows where(
    as Column( name ) == "**skipped**" | As Column( name ) ==
    "dne"
   ))
  ),
  Insert Into( colsToDelete, name )
 )
);

dt << delete columns( colsToDelete );

回复：用于选择包含值列表的所有列的脚本？

根据您的数据和所需的速度，您可能想要更改获取唯一值列表的方式，但以下是一种选择

Names Default To Here(1);

dt = New Table("Untitled 4",
 Add Rows(3),
 Compress File When Saved(1),
 New Column("Column 1", Character, "Nominal", Set Values({"dne", "**skipped**", "**skipped**"})),
 New Column("Column 2", Character, "Nominal", Set Values({"dne", "a", "b"})),
 New Column("Column 3", Character, "Nominal", Set Values({"**skipped**", "c", "d"})),
 New Column("Column 4", Character, "Nominal", Set Values({"dne", "dne", "**skipped**"}))
);

aa_colmask = ["dne" => 1, "**skipped**" => 1]; // we don't need the values, just keys

cols_to_remove = {};

For Each({colname}, dt << Get Column Names("String"),
 // depending on how much data you have, you might want to use different unique value techniques
 // Summarize, Associative Array and Summary are few options.
 // You can also first check if there are exactly 2 unique values, then check those
 aa_values = Associative Array(Column(dt, colname)); // I use AA here as it is the easiest option (not fastest!)
 aa_values << Remove(aa_colmask);
 If(N Items(aa_values) == 0,
  Insert Into(cols_to_remove, colname);
 );
);

dt << Delete Columns(cols_to_remove);