aGrUM 2.3.2
a C++ library for (probabilistic) graphical models
rawDatabaseTable.h File Reference

The table containing the raw/original data of a database. More...

Include dependency graph for rawDatabaseTable.h:
This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Classes

class  gum::learning::RawDatabaseTable
 The table containing the raw/original data of a database. More...

Namespaces

namespace  gum
 gum is the global namespace for all aGrUM entities
namespace  gum::learning
 include the inlined functions if necessary

Detailed Description

The table containing the raw/original data of a database.

Class RawDatabaseTable is intended to store in RAM the raw/original data of a database. Such raw data are not well suited for learning tasks because they need to be interpreted by the learning algorithm, which would incur a strong overhead. However, reading a CSV file and interpreting its data in order to reshape them in a way that will allow fast parsing by learning algorithms is also very time consuming. So, if you are unsure about the correct interpretation and need to change it several times either before processing the learning or during several learning phases, it is efficient to first read the CSV file and store its useful data (removing comments, for instance) into a first database table and, then, use this preprocessed table to produce quickly the interpreted database table that will subsequently be used by the learning. The purpose of the RawDatabaseTable class is precisely to implement this preprocessed table.

Author
Christophe GONZALES(_at_AMU) and Pierre-Henri WUILLEMIN(_at_LIP6)

Definition in file rawDatabaseTable.h.