com.intel.analytics.bigdl.dataset.datamining
A RowTransformer which transform each selected columns to a size(1) Tensor.
A RowTransformer which transform each selected columns to a size(1) Tensor.
The keys of output Table are indices of selected columns.
indices of selected columns
size of Row transformed by this transformer
A RowTransformer which transform each selected columns to a size(1) Tensor.
A RowTransformer which transform each selected columns to a size(1) Tensor.
The keys of output Table are fieldNames of selected columns.
field names of selected columns
A RowTransformer which contains both atomic schemas and numeric schemas.
A RowTransformer which contains both atomic schemas and numeric schemas.
field names of selected columns
Map<schemaKey, fieldNames of selected columns> of numeric fields
A RowTransformer which concat values of selected columns to one Tensor.
A RowTransformer which concat values of selected columns to one Tensor.
It means you will get a Table with keys of numericFields.
Values of Table are Tensors concatenated by selected columns of the keys.
Map<schemaKey, fieldNames of selected columns> of numeric fields
A RowTransformer which concat values of all columns to one Tensor.
A RowTransformer which concat values of all columns to one Tensor.
It means you will get a Table with single key-value pair after transformation.
The unique key is schemaKey. The unique value is a size(length of Row) Tensor.
key of the schema, default value is "all"