Example of processing configuration file

The following code shows the available configuration and all the default values. Remember that you can always add your own variables and you will find them in the DataProcessor object!

# Directory to save the logs files, level for logging in the console (console_level) and
# in the log file (file_level)
log:
  file_path: ./musif.log
  file_level: DEBUG
  console_level: ERROR

# Flag to determine whether to delete rows associated to files containing errors (whose
# FileNames bust be contained in a './errors.csv' file) or not
delete_files: False

# Flag to determine whether delete rows associated to files that do not have harmonic
# annotations (and
# hence, no harmonic analysis) or not
delete_files_without_harmony: False

# Flag to determine if multiple voices (e.g. duetos and trios) sould be merge into
# the same column (SoundVoice_*), calculating averages, min or max of both values 
merge_voices: True

# Flag to determine if keys, keys modulatory and degrees columns should be grouped per type
# For instance, about the keys:
# * IV, II, and VI will be merged into the SD group (subdominant)
# * iv, ii, will be merged into the sd group (subdominant)
# * I and i will be merged into the T group (tonic)
# * III and vi will be merged into the rel group (relative)
# 
# The degrees will be merged into [Doubly] Augmented, [Doubly] Diminished, Natural, No
# Natural. This applies to both raw and percentage degree features.
grouped: False

# Flag to determine if Intrumentation column should be disgregated in several columns
# (Presence_of_*) having a value of 1 if instrument is present, 0 otherwise.
# see also `delete_presence` option
separate_intrumentation_column: False

# Specify the instruments whose features will be kept even if they are merged with the option
# `grouped`
# Matching is case-sensitive
instruments_to_keep: []
  # Possible values are from abbreviations from musif.musicxml.constants.SOUND_TO_ABBREVIATION
  # Note that if an abbreviation is not available, the full instrument name is used
  # Example:
  # - vnI
  # - bs
 
# Specify the instuments whose features will be deleted
# i.e. 
# * features containing 'Part' + <inst> will be deleted
# * features containing <inst> + '_' will be deleted
# Matching is case-sensitive
instruments_to_delete: []
  # Possible values are from abbreviations from musif.musicxml.constants.SOUND_TO_ABBREVIATION
  # Note that if an abbreviation is not available, the full instrument name is used
  # First letter is always capitalized
  # Example:
  # - Ob
  # - Cor
  # - Bn
  # - Va
  # - Eh
  # - Tbn
  # - TbI
  # - Timp
  # - Tr
  # - TrI
  # - BsI
  # - BsII
  # - VeI
  # - VeII
  # - CI
  # - CiII
  # - CiI

# Rows having more than this percentage of NaN will be deleted
# value null (or None) means "keep all"
max_nan_rows: null

# Columns having more than this percentage of NaN will be deleted (applied after
# `max_nan_rows`)
# value null (or None) means "keep all"
max_nan_columns: null

# Columns containing these substrings will have nan values replaced with 0
# Applied after `max_nan_rows` and `max_nan_columns`
# Matching is case-sensitive
replace_nans: []
  # Example of values (but any substring can be used):
  # - Interval
  # - Degree
  # - Harmony

# Columns ending with any of these strings will be deleted
# Matching is case-sensitive
columns_endswith: []
  # example of values (but any other suffix can be used)
  # - _HighestNote
  # - _LowestNote
  # - _Notes
  # - _SoundingMeasures
  # - _Syllables
  # - _NumberOfFilteredParts
  # - NumberOfParts
  # - _NotesMean
  # - Librettist
  # - _LargestIntervalAsc
  # - _LargestIntervalAll
  # - _LargestIntervalDesc
  # - _NotesMean
  # - 'Semitones_Sum'    
  # - _MeanInterval'

# Columns starting with any of these strings will be deleted
# Matching is case-sensitive
columns_startswith: []
  # example of values (but any other prefix can be used)
  # - FamilyWw
  # - FamilyBr
  # - FamilyGen
  # - FamilyGen
  # - FamilyPerc
  # - EndOfThemeA
  # - NumberOfBeats
  # - SoundVoice_Dyn

# Columns containing any of these strings will be deleted
# Matching is case-sensitive
columns_contain: []
  # example of values (but any other substring can be used)
  # - _Count
  # - _SoundingMeasuresMean
  # - _SmallestSemitones
  # - _SmallestAbsolute
  # - _SmallestInterval
  # - Bn
  # - VnII
  # - VnIII
  # - VnIV
  # - Vc
  # - Cl
  # - Tpt
  # - Hn
  # - Fl

# Columns exactly matching any of these strings will be deleted
# Matching is case-sensitive
columns_match: []
  # example of values (but any column can be used)
  # - Presence_of_vn
  # - Presence_of_bs
  # - Presence_of_vc
  # - Presence_of_va