You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In the ngram indexer, at present only the 'vertical' setting allows 'all' as input, despite that in the examples in the documentation it implies that 'all' could also be passed as an argument to the 'horizontal' setting (see lines 296-298).
The larger bug within this bug is that the horizontal arguments at present only allow 'upper' or 'lower' but cannot be both. Even if you manually type in a list of the necessary horizontal arguments, the ngrams can only be properly calculated for one column of data (i.e. one vertical setting or pair of voices). If you try to pass it multiple columns in the 'vertical' setting and then calculate both the upper and lower horizontal intervals by manually passing a list, it outputs the ngrams incorrectly (see code below for clarification.)
##Code for single column, which calculates both sets of melodic intervals in the ngram:
In the ngram indexer, at present only the 'vertical' setting allows 'all' as input, despite that in the examples in the documentation it implies that 'all' could also be passed as an argument to the 'horizontal' setting (see lines 296-298).
The larger bug within this bug is that the horizontal arguments at present only allow 'upper' or 'lower' but cannot be both. Even if you manually type in a list of the necessary horizontal arguments, the ngrams can only be properly calculated for one column of data (i.e. one vertical setting or pair of voices). If you try to pass it multiple columns in the 'vertical' setting and then calculate both the upper and lower horizontal intervals by manually passing a list, it outputs the ngrams incorrectly (see code below for clarification.)
##Code for single column, which calculates both sets of melodic intervals in the ngram:
ngram_settings = {
'n': 2,
'vertical': [('0,1',)],
'horizontal': [('0','1',)],
'brackets': True
}
ng = my_score.get_data('ngram', data=[hint, mint],
settings=ngram_settings)
##Code for multiple columns, which outputs the following:
ngram_settings = {
'n': 2,
'vertical': [('0,1','0,3')],
'horizontal': [('0','1','0','3')],
'brackets': True
}
The text was updated successfully, but these errors were encountered: