Skip to content

Instantly share code, notes, and snippets.

@felipessalvatore

felipessalvatore/ngram_overlap.py

Created December 18, 2017 17:00
Show Gist options
  • Save felipessalvatore/f6edad81731c6559ba0aa537f94277b7 to your computer and use it in GitHub Desktop.
Save felipessalvatore/f6edad81731c6559ba0aa537f94277b7 to your computer and use it in GitHub Desktop.
Download ZIP
using nltk to count N-gram overlap between two sentences
Raw
ngram_overlap.py
from nltk.translate.bleu_score import modified_precision
pt = "em plano aberto, a cidade parece linda"
ref = "in a wide shot, the city looks beautiful"
c1 = "in the open, the city looks beautiful"
c2 = "in open plan, the city looks gorgeous"
reference = [ref.split(" ")]
candidate1 = c1.split(" ")
candidate2 = c2.split(" ")
for i in range(4):
print("{}-gram".format(i+1))
print(modified_precision(reference, candidate1, i+1))
print()
for i in range(4):
print("{}-gram".format(i+1))
print(modified_precision(reference, candidate2, i+1))
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment