203 lines (183 with data), 6.8 kB

##########################
# CONFIGURATION FILE FOR VARIABLES
# USED ACROSS THE PROJECT
##########################


##########################  PIPELINE  ##########################
# Pipeline flags for what processes to perform on the data
pipeline:
  # What  to read
  in:
    # What's the source of the corpus.
    source: mongo # 'mongo', 'file' and 'delete' values currently supported
    # What's the type of the input to be read. Will use
    # the corresponding fields in this setting file.
    type: edges # 'text', 'edges' and 'med_rec' currently
    # Do we want to do it all in a streaming fashion?
    stream: False
    # Do we want to perform parallel semrep processing?
    # Currently for semrep extraction and neo4j saving
    parallel: True
  # What to do with it
  trans:
    # Extract entities using metamap?
    metamap: False
    # Extract relations using reverb?
    reverb: False
    # Extract entities + relations through semrep?
    semrep: True
    # Transform edges, fetching concepts and other info?
    get_concepts_from_edges: False
  # What to do with the outcome
  out:
    # Create json output?
    json: False
    # Create .csv files for neo4j?
    csv: False
    # Create-update neo4 db?
    neo4j: True
    # Create-update sentences in mongo?
    mongo_sentences: False
    # Create-update collection of documents/edges in mongo?
    mongo: False
########################## END PIPELINE ##########################





##########################  INPUT ##########################
# Variables regarding loading
load:
  # Paths to binaries/files
  path:
    # Path to metamap
    metamap: /media/kostas/DATA/LLD/Metamap/public_mm/bin/metamap16
    # Path to reverb
    reverb: /media/kostas/DATA/GIT/reverb/core/
    # Path to semrep
    semrep: /media/kostas/DATA/LLD/SEMREP/public_semrep/bin/
    # Path toAscii binary
    toAscii: /media/kostas/DATA/LLD/Papers/BioASQ/MARIOS_PROJECT/lvg2017/bin
    # Path to input File
    file_path: /media/kostas/DATA/LLD/Papers/BioASQ/MARIOS_PROJECT/enriched_doid.json
  # Mongo input variables
  mongo:
    # DB Full uri. If user/pass required pass it here like mongodb://user:pass@host:port
    uri: mongodb://0.0.0.0:27017
    # DB name
    db: iasis_20190620_KGupdate
    # Collection name
    collection: LC_20190620_pubmed_MeSH_ENRICHED
    # Cache collection name
    cache_collection: cache
    # FOR PRINTING PURPOSES ONLY!
    file_path: mongodb://localhost:27017/iasis_20190620_KGupdate|LC_20190620_pubmed_MeSH_ENRICHED
  # For medical records
  med_rec:
    # Medical record name of the column to read text from
    textfield: PrimaryDiagnosisDescription
    # Field separator
    sep: \t
    # idfield
    idfield: AdmissionID
  # For article-document type of input
  text:
    # Outer field name for the documents residing in a json probably
    itemfield: documents
    # Name of the field to read text from
    textfield: abstractText #body_Filtered
    # Idfield for each document
    idfield: pmid
    # Label field for each article
    labelfield: title
    # Sentence Prefix (is this abstract or fullText)
    sent_prefix: abstract
  # For relation-edge type of input
  edges:
    # Name of the field where edges are stored
    itemfield: relations
    # Type of the subject
    # Currently [Entity, Article and any new one]
    sub_type: Article
    # Type of the object
    obj_type: Entity
    # Id_field of the subject
    # Currently [UMLS for cuis, PMC or TEXT for articles, None for precomputed and whatever else for DRUGBANK,MSH etc]
    sub_source: None
    # Id_field of the subject
    obj_source: None
##########################  END INPUT ##########################



##########################  API KEYS ##########################
# Variables regarding apis
apis:
  # Biontology for mapping cuis to uris
  biont: ********-****-****-****-************
  # Umls ticket service
  umls: ********-****-****-****-************
##########################  END API KEYS ##########################

##########################  NEO4j ##########################
# Neoj variables
neo4j:
  # DB url
  host: localhost
  # port
  port: 7474
  # user name
  user: neo4j
  # password
  password: admin
  # name of the resource to be inputted in neo4j
  resource: text
##########################  END NEO4j ##########################


##########################  MONGODB FOR SENTENCES ##########################
# Mongo variables for writing sentences
mongo_sentences:
  # DB Full uri. If user/pass required pass it here like mongodb://user:pass@host:port
  uri: mongodb://localhost:27017
  # DB name
  db: Marios
  # Collection name
  collection: Articles2
##########################  END MONGODB ##########################


########################## CACHE  ############################
cache_path: /media/kostas/DATA/LLD/Papers/BioASQ/MARIOS_PROJECT/cache.json
########################## END CACHE  ############################

########################## LOG  ############################
log_path: /media/kostas/DATA/LLD/Papers/BioASQ/MARIOS_PROJECT/medknow.log
########################## END LOG  ############################


########################## PARALLEL  ############################
# Number of cores to use in parallel schemes. If none, it will be
# equal to cpu count, using all available cores.
num_cores: 4
# Number of items per core to be processed. This will create a batch
# of total size = num_cores*batch_per_core. It defaults to 100
batch_per_core: 100
########################## END PARALLEL  ############################

##########################  OUTPUT ##########################
# Output variables
out:
   # Resulting .json file before neo4j
  json:
    # Path
    out_path: /media/kostas/DATA/LLD/Papers/BioASQ/MARIOS_PROJECT/MARIOS_ERRORS/small2.json
    # Output json outer field name for the documents
    itemfield: relations
    # Output json text field for each document
    json_text_field: text
    # Output Idfield
    json_id_field: id
    # Output Label field
    json_label_field: title
  # Resulting .json file before neo4j
  csv:
    # Path
    out_path: /media/kostas/DATA/LLD/Papers/BioASQ/MARIOS_PROJECT/out
  # Resulting .json file before neo4j
  neo4j:
    # Just for printing! Change the Neo4j field variables, not this!
    out_path: localhost:7474
  mongo:
    # DB Full uri. If user/pass required pass it here like mongodb://user:pass@host:port
    uri: mongodb://localhost:27017
    # DB name
    db: iasis_20190620_KGupdate
    # Collection name
    collection: LC_20190620_pubmed_MeSH_ENRICHED
    # Just for printing! Change the mongo field variables, not this!
    out_path: localhost:27017/iasis_20190620_KGupdate|LC_20190620_pubmed_MeSH_ENRICHED
##########################  END OUTPUT ##########################