Reputation: 363
Using nltk to generate a list of synonyms to based on an input list of keywords.
I'm getting "TypeError: 'NoneType' object is not iterable" when nltk doesn't have a synonym (i.e., when "None" is returned).
from nltk.corpus import stopwords
from PyDictionary import PyDictionary
dictionary=PyDictionary()
inputWords = ['word1','word2','word3','word4','word5','word6','word7','word8']
synonyms = list(dictionary.synonym(i) for i in inputWords
current output for synonyms:
[['syn1','syn2','syn3'],['syn4'],None,['syn5'],None,['syn6','syn7'],None,['syn8'],None]]
desired output for synonyms:
['syn1','syn2','syn3','syn4','syn5','syn6','syn7','syn8']
I tried:
flat_list = [item for sublist in synonyms for item in sublist]
The output:
TypeError: 'NoneType' object is not iterable
What do I write to return just the synonyms in a clean list? I prefer to use list comprehension. Any help appreciated. I'm a python noob and can't find the exact answer.
Upvotes: 0
Views: 635
Reputation: 95873
A pythonic way to do this:
result = []
for word in input_words:
synonyms = dictionary.synonym(word)
if synonyms:
result.extend(synonyms)
If you insist on using a list comprehension, you could use an assignment expression:
[synonym for word in input_words if (synonyms:=dictionary.synonym(words)) for synonym in synonyms]
But I think the first option is much preferable.
Upvotes: 0
Reputation: 44
Try:
flat_list = []
for i in range(len(synonyms):
if isinstance(synonyms, list):
flat_list += i
Upvotes: 1
Reputation: 1116
Here is a more pythonic one liner:
flat_list = [i for sublist in synonyms if sublist is not None for i in sublist]
flat_list
>>>
['syn1', 'syn2', 'syn3', 'syn4', 'syn5', 'syn6', 'syn7', 'syn8']
Upvotes: 1
Reputation: 6181
You can remove the None
values and use itertools -
import itertools
list(itertools.chain(*[x for x in synonyms if x]))
Upvotes: 2