Skip to content
This repository has been archived by the owner on Nov 3, 2023. It is now read-only.

AcceptabilityChecker min words adjustable #3846

Merged
merged 3 commits into from
Jul 23, 2021
Merged

AcceptabilityChecker min words adjustable #3846

Changes from 2 commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
c3d89da
min words as a param
mojtaba-komeili May 25, 2021
f093d96
min words as input param
mojtaba-komeili Jul 22, 2021
e862ff4
pr comment
mojtaba-komeili Jul 23, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 6 additions & 2 deletions parlai/crowdsourcing/utils/acceptability.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,9 @@
from parlai.utils.safety import OffensiveStringMatcher


DEFAULT_MIN_WORDS_THRESHOLD = 3


class AcceptabilityChecker:

ALL_VIOLATION_TYPES = [
Expand All @@ -20,8 +23,9 @@ class AcceptabilityChecker:
'safety',
]

def __init__(self):
def __init__(self, min_words=DEFAULT_MIN_WORDS_THRESHOLD):
self.offensive_lang_detector = OffensiveStringMatcher()
self.min_words_violation_threshold = min_words

def check_messages(
self,
Expand Down Expand Up @@ -57,7 +61,7 @@ def check_messages(
# Do messages have the minimum acceptable average number of words?
if 'min_words' in violation_types:
total_num_words = sum([len(message.split()) for message in messages])
if total_num_words / len(messages) < 3:
if total_num_words / len(messages) < self.min_words_violation_threshold:
violations.append('under_min_length')

# Does the first message start with a greeting, indicating that the Turker
Expand Down