Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/exegia/corpora-py/llms.txt

Use this file to discover all available pages before exploring further.

describe_feature() gives you detailed information about a single feature: its stored metadata (such as a human-readable description and value type) and its most frequent values ranked by occurrence count. Use this tool when you need to know what values a feature can take before writing a search template, or when you want to validate that a feature name is correct and understand its data distribution.

​
Parameters

​
feature
string
required
The name of the feature to inspect (e.g. "lex", "pos", "gloss"). Must exactly match a feature present in the corpus. Use list_features() to discover valid names.
​
sample_size
integer
How many top values to include in the frequency list, ordered from most to least common. Defaults to 20.
​
corpus
string
Corpus name. When omitted, the currently active corpus is used.

​
Returns

A multi-line formatted string with three sections:
  • Feature — the name of the feature as requested.
  • Metadata — key/value pairs drawn from the feature’s stored metadata (e.g. description, valueType). Shows (none) when no metadata is stored.
  • Top N values — a frequency table of the sample_size most common values, with each value left-aligned in a 30-character column and its occurrence count right-aligned with comma formatting.
If the feature name is not found, returns "Feature '<name>' not found.".

​
Example

result = describe_feature(feature="lex", sample_size=5)
print(result)
Expected output: 
Feature: lex
Metadata:
  description: Lexeme
  valueType: str

Top 5 values (value / count):
  'H'                              8,632
  'W'                              5,421
  'B'                              4,804
  'L'                              3,927
  'MN'                             2,510

Build docs developers (and LLMs) love

Get started for freeTalk to us