update README
Browse files- distinct.py +3 -3
distinct.py
CHANGED
@@ -52,10 +52,11 @@ _CITATION = """\
|
|
52 |
|
53 |
_DESCRIPTION = """\
|
54 |
Distinct metric is to calculate corpus-level diversity of language. We provide two versions of distinct score. Expectation-Adjusted-Distinct is the default one, which removes
|
55 |
-
the biases of the original distinct score on
|
56 |
|
57 |
For the use of Expectation-Adjusted-Distinct, vocab_size is required.
|
58 |
-
|
|
|
59 |
"""
|
60 |
|
61 |
|
@@ -96,7 +97,6 @@ BAD_WORDS_URL = "http://url/to/external/resource/bad_words.txt"
|
|
96 |
class distinct(evaluate.Measurement):
|
97 |
|
98 |
def _info(self):
|
99 |
-
# TODO: Specifies the evaluate.EvaluationModuleInfo object
|
100 |
return evaluate.MeasurementInfo(
|
101 |
# This is the description that will appear on the modules page.
|
102 |
module_type="measurement",
|
|
|
52 |
|
53 |
_DESCRIPTION = """\
|
54 |
Distinct metric is to calculate corpus-level diversity of language. We provide two versions of distinct score. Expectation-Adjusted-Distinct is the default one, which removes
|
55 |
+
the biases of the original distinct score on lengthier sentences. Distinct is the original version.
|
56 |
|
57 |
For the use of Expectation-Adjusted-Distinct, vocab_size is required.
|
58 |
+
|
59 |
+
Please follow ACL paper https://aclanthology.org/2022.acl-short.86 for motivation and follow the rules of thumb provided by https://github.com/lsy641/Expectation-Adjusted-Distinct/blob/main/EAD.ipynb to determine the vocab_size
|
60 |
"""
|
61 |
|
62 |
|
|
|
97 |
class distinct(evaluate.Measurement):
|
98 |
|
99 |
def _info(self):
|
|
|
100 |
return evaluate.MeasurementInfo(
|
101 |
# This is the description that will appear on the modules page.
|
102 |
module_type="measurement",
|