fix bug when setting stop_words to None

TimSchopf · TimSchopf · commit 3dc3212ee865 · 2024-05-02T15:22:13.000+02:00
Signed-off-by: Tim Schopf &lt;tim.schopf@t-online.de&gt;
diff --git a/keyphrase_vectorizers/keyphrase_vectorizer_mixin.py b/keyphrase_vectorizers/keyphrase_vectorizer_mixin.py
@@ -348,7 +348,7 @@ def _get_pos_keyphrases(self, document_list: List[str], stop_words: Union[str, L
             )
 
 
-        stop_words_list = []
+        stop_words_list = set()
         if isinstance(stop_words, str):
             try:
                 stop_words_list = set(nltk.corpus.stopwords.words(stop_words))

Original file line number	Diff line number	Diff line change
`@@ -348,7 +348,7 @@ def _get_pos_keyphrases(self, document_list: List[str], stop_words: Union[str, L`
`348`	`348`	`)`
`349`	`349`
`350`	`350`
`351`		`- stop_words_list = []`
	`351`	`+ stop_words_list = set()`
`352`	`352`	`if isinstance(stop_words, str):`
`353`	`353`	`try:`
`354`	`354`	`stop_words_list = set(nltk.corpus.stopwords.words(stop_words))`