N1904addons - Feature: ms{ind}_lem_base_bc

Feature group	Feature type	Data type	Available for node types	Feature status
Morpheus	Node	int	word	✅

Feature description

Summary feature for grouped analysis #{ind} providing the base lemma (clean; without suffixes) encoded in uniccode.

This is a Morpheus summary data feature.

Feature values

The lemma in unicode.

Coding

The following Python code demonstrates how to programaticaly obtain details like lemma and morphological tags per grouped analysis blocks.

wordNodes = F.otype.s("word")
for wordNode in wordNodes:
    for blockNumber in range(1, 9):
        # dynamically get F.ms{blockNumber}_lemma & morph
        lemma = Fs(f"ms{blockNumber}_lemma").v(wordNode)
        if not lemma: continue
        morph_string = Fs(f"ms{blockNumber}_morph").v(wordNode)
        morph_sim_string = Fs(f"ms{blockNumber}_morph_sim").v(wordNode)
        # decompose on the slash
        parts = morph_string.split("/")
        # print what was found
        print(f"node={wordNode}, number={blockNumber} → lemma={lemma}, tags: {parts}")

node=1, number=1 → lemma=Βίβλος, tags: ['N-NSM']
node=1, number=2 → lemma=βίβλος, tags: ['N-NSF']
node=2, number=1 → lemma=γένεσις, tags: ['N-GSF']
node=3, number=1 → lemma=Ἰησοῦς, tags: ['N-GSM', 'N-VSM', 'N-PRI']
...

The snippet below uses a list comprehension to dynamically generate the names of Morpheus features indexed by number. This allows the resulting list to be passed directly to A.show() for displaying multiple feature layers at once.

# Dynamically generate feature names for all morphology sets
morphFeatureList = (
    [f'ms{ind}_lemma'      for ind in range(1, 9)]
  + [f'ms{ind}_morph'      for ind in range(1, 9)]
  + [f'ms{ind}_morph_sim'  for ind in range(1, 9)]
)
# Display the query results with the Morpheus features
A.show(QueryResult, extraFeatures=morphFeatureList)

The image below shows a syntax tree with the display of these features enabled.

Data source

GitHub repository Create_morpheus_TF_dataset.