Sentence complete classifier #389
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| @@ -0,0 +1 @@ | |||
| Languages can be very dynamic and complicated. This brick does not actually try to be able to accurately classify all sentences, which would be quite complex. Instead, this brick is meant to check if some characteristics apply that a lot of complete sentences have. These characteristics being: does the sentence starts with an uppercase character, if it ends on a punctuation and if it contains at least two nouns and a verb. The name `starts_with_uppercase_ends_with_punctuation_and_contains_two_nouns_and_a_verb` would be a bit long for a brick, though. | |||
There was a problem hiding this comment.
From the name and description of the brick, I would not expect the aggregation logic.
I suggest mentioning it in the README
Comment on lines
+21
to
+24
| cognition_init_mapping = { | ||
| "incomplete": "Needs fix", | ||
| "complete": "null" | ||
| }, |
There was a problem hiding this comment.
the mapping for partly complete is missing
| "incomplete": "Needs fix", | ||
| "complete": "null" | ||
| }, | ||
| integrator_inputs={ |
There was a problem hiding this comment.
outputs are missing
Comment on lines
+21
to
+36
| classifications = [] | ||
| for sent in doc.sents: | ||
| if sent[0].is_title and sent[-1].is_punct: | ||
| has_noun = 2 | ||
| has_verb = 1 | ||
| for token in sent: | ||
| if token.pos_ in ["NOUN", "PROPN", "PRON"]: | ||
| has_noun -= 1 | ||
| elif token.pos_ == "VERB": | ||
| has_verb -= 1 | ||
| if has_noun < 1 and has_verb < 1: | ||
| classifications.append("complete") | ||
| else: | ||
| classifications.append("incomplete") | ||
| else: | ||
| classifications.append("incomplete") |
There was a problem hiding this comment.
What do you think of the following restructuring of the code?
classifications = []
for sent in doc.sents:
if not (sent[0].is_title and sent[-1].is_punct):
classification.append("incomplete")
continue
has_noun = 2
has_verb = 1
for token in sent:
if token.pos_ in ["NOUN", "PROPN", "PRON"]:
has_noun -= 1
elif token.pos_ == "VERB":
has_verb -= 1
if has_noun < 1 and has_verb < 1:
classifications.append("complete")
continue
classifications.append("incomplete")
another option could also be to encapsulate the per sentence classification in a function, and call this via a list comprehension
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
refinery
API
common code
additional points:
__init__.py)@FelixKirschKern Not sure at all about the logic behind this brick, would love to get your feedback. Classifying whether or not a sentence is complete is not easy, but I think it's alright to check if a sentence already has some features we would expect in a usual sentence (Uppercase character in the beginning, end with a punctuation and contains nouns and a verb). This will of course miss some sentences. I also expect that the input will rarely just be one sentence, but multiple sentences in a text that might just be cut off at the end due to chunking. There might be better ways for for the aggregation part, too.