.. _fwflag_add_lex_table: =============== --add_lex_table =============== Switch ====== --add_lex_table [-l LEX] Description =========== Add a lexicon-based feature table Argument and Default Value ========================== --add_lex_table -l (where is the name of the table) Details ======= Given an existing N-gra,s table as your original feature table, this switch produces a new feature table where the features are now topics (as opposed to N:doc:`fwflag_grams`) based on the topics in the LEX table. Unless :doc:`fwflag_f` is provided, the table feat$1gram$TABLE$GROUP_BY_FIELD$16to16 must be present. Other Switches ============== Required Switches: * :doc:`fwflag_d`, :doc:`fwflag_c`, :doc:`fwflag_t`, :doc:`fwflag_l` Optional Switches: * :doc:`fwflag_f` [alternative n-gram table; can be used in place of word table] * :doc:`fwflag_weighted_lexicon` :doc:`fwflag_anscombe`, :doc:`fwflag_sqrt`, :doc:`fwflag_log`, :doc:`fwflag_boolean` (transformations, default is no transformation) Example Commands ================ .. code-block:: python # General form # Creates the table: feat$cat_LEX_TABLE$TABLE$GROUP_BY_FIELD$16to# # In the above # = (1, 2, 4, 8, 16) and depends on the transformation (boolean, log, sqrt, anscombe and none, respectively) ./dlatkInterface.py -d DATABSE -t TABLE -c GROUP_BY_FIELD -add_lex_table -l LEX_TABLE # Creates the table: feat$cat_met_a30_2000_cp_w$primals_new$dp_id$16to16 ./dlatkInterface.py -d primals -t primals_new -c dp_id --group_freq_thresh 40000 --add_lex_table -l met_a30_2000_cp --weighted_lex