diff --git a/concise_concepts/conceptualizer/Conceptualizer.py b/concise_concepts/conceptualizer/Conceptualizer.py index 173c7fb..3877f16 100644 --- a/concise_concepts/conceptualizer/Conceptualizer.py +++ b/concise_concepts/conceptualizer/Conceptualizer.py @@ -64,8 +64,6 @@ def __init__( if ent_score: Span.set_extension("ent_score", default=None) self.ent_score = ent_score - self.orignal_words = [j for i in data.values() for j in i] - self.original_data = deepcopy(data) self.data = data self.name = name self.nlp = nlp @@ -252,15 +250,13 @@ def verify_data(self, verbose: bool = True) -> None: else: raise Exception(msg) self.data = deepcopy(verified_data) + self.original_data = deepcopy(verified_data) def expand_concepts(self) -> None: """ For each key in the data dictionary, find the topn most similar words to the key and the values in the data dictionary, and add those words to the values in the data dictionary """ - - self.original_data = deepcopy(self.data) - for key in self.data: present_key = self._check_presence_vocab(key) if present_key: diff --git a/pyproject.toml b/pyproject.toml index 352afd8..df64304 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -1,6 +1,6 @@ [tool.poetry] name = "concise-concepts" -version = "0.7.4" +version = "0.7.4.1" description = "This repository contains an easy and intuitive approach to few-shot NER using most similar expansion over spaCy embeddings. Now with entity confidence scores!" authors = ["David Berenstein "] license = "MIT"