:py:mod:`medcat.utils.model_creator`
====================================

.. py:module:: medcat.utils.model_creator


Module Contents
---------------


Functions
~~~~~~~~~

.. autoapisummary::

   medcat.utils.model_creator.create_cdb
   medcat.utils.model_creator.create_vocab
   medcat.utils.model_creator.train_unsupervised
   medcat.utils.model_creator.create_models
   medcat.utils.model_creator.main



Attributes
~~~~~~~~~~

.. autoapisummary::

   medcat.utils.model_creator.DEFAULT_UNIGRAM_TABLE_SIZE
   medcat.utils.model_creator.logger
   medcat.utils.model_creator.parser


.. py:data:: DEFAULT_UNIGRAM_TABLE_SIZE
   :value: 100000000

   

.. py:data:: logger

   

.. py:function:: create_cdb(concept_csv_file, medcat_config)

   Create concept database from csv.

   :param concept_csv_file: Path to CSV file containing all concepts and synonyms.
   :type concept_csv_file: Path
   :param medcat_config: MedCAT configuration file.
   :type medcat_config: Config

   :Returns: **CDB** -- MedCAT concept database containing list of entities and synonyms, without context embeddings.


.. py:function:: create_vocab(cdb, training_data_list, medcat_config, output_dir, unigram_table_size)

   Create vocabulary for word embeddings and spell check from list of training documents and CDB.

   :param cdb: MedCAT concept database containing list of entities and synonyms.
   :type cdb: medcat.cdb.CDB
   :param training_data_list: List of example documents.
   :type training_data_list: list
   :param medcat_config: MedCAT configuration file.
   :type medcat_config: medcat.config.Config
   :param output_dir: Output directory to write vocabulary and data.txt (required to create vocabulary) to.
   :type output_dir: pathlib.Path
   :param unigram_table_size: Size of unigram table to be initialized before creating vocabulary.
   :type unigram_table_size: int

   :Returns: **medcat.vocab.Vocab** -- MedCAT vocabulary created from CDB and training documents.


.. py:function:: train_unsupervised(cdb, vocab, config, output_dir, training_data_list)

   Perform unsupervised training and save updated CDB.

   Although not returned explicitly in this function, the CDB will be updated with context embeddings.

   :param cdb: MedCAT concept database containing list of entities and synonyms.
   :type cdb: medcat.cdb.CDB
   :param vocab: MedCAT vocabulary created from CDB and training documents.
   :type vocab: medcat.vocab.Vocab
   :param config: MedCAT configuration file.
   :type config: medcat.config.Config
   :param output_dir: Output directory to write updated CDB to.
   :type output_dir: pathlib.Path
   :param training_data_list: List of example documents.
   :type training_data_list: list

   :Returns: **medcat.cdb.CDB** -- MedCAT concept database containing list of entities and synonyms, as well as context embeddings.


.. py:function:: create_models(config_file)

   Create MedCAT CDB and Vocabulary models.

   :param config_file: Location of model creator configuration file to specify input, output and MedCAT configuration.
   :type config_file: pathlib.Path

   :Returns: **medcat.cat.CAT** -- Containing CDB, Vocab and Config.


.. py:function:: main(config_file)


.. py:data:: parser