Spaces:
Paused
Paused
Commit
•
988ba41
1
Parent(s):
fc0306d
more improvements to chatgpt audit
Browse files- audits/1720097343.csv +1 -0
- audits/1720097580.csv +1 -0
- chatgpt_audit.py +4 -1
audits/1720097343.csv
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
input_word,original_dictionary_word,new_dictionary_word
|
audits/1720097580.csv
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
input_word,original_dictionary_word,new_dictionary_word
|
chatgpt_audit.py
CHANGED
@@ -1,3 +1,4 @@
|
|
|
|
1 |
import os
|
2 |
import csv
|
3 |
import json
|
@@ -32,7 +33,9 @@ model_name = 'sentence-transformers/all-mpnet-base-v2'
|
|
32 |
model = SentenceTransformer(model_name)
|
33 |
|
34 |
def remove_misc_words(text):
|
35 |
-
|
|
|
|
|
36 |
|
37 |
def compare_embeddings(old_dictionary_word, new_dictionary_word):
|
38 |
old_dictionary_word = remove_misc_words(old_dictionary_word)
|
|
|
1 |
+
import re
|
2 |
import os
|
3 |
import csv
|
4 |
import json
|
|
|
33 |
model = SentenceTransformer(model_name)
|
34 |
|
35 |
def remove_misc_words(text):
|
36 |
+
text = text.replace('raw', '').replace('fresh', '').replace('canned', '').replace('prepared', '').replace('with', '').replace('and', '').replace('or', '').replace('in', '').replace('NFS', '').strip()
|
37 |
+
text = re.sub(r'\(.*?\)', '', text)
|
38 |
+
return text
|
39 |
|
40 |
def compare_embeddings(old_dictionary_word, new_dictionary_word):
|
41 |
old_dictionary_word = remove_misc_words(old_dictionary_word)
|