Commit
·
e4bc2b6
1
Parent(s):
eba1c32
Update app.py
Browse files
app.py
CHANGED
|
@@ -1,12 +1,15 @@
|
|
| 1 |
import torch
|
| 2 |
-
import nltk
|
| 3 |
import streamlit as st
|
| 4 |
from transformers import RobertaTokenizer, RobertaForSequenceClassification
|
|
|
|
|
|
|
|
|
|
| 5 |
nltk.download('stopwords')
|
| 6 |
nltk.download('punkt')
|
| 7 |
stop_words = set(stopwords.words('english'))
|
| 8 |
stop_words.discard('and')
|
| 9 |
|
|
|
|
| 10 |
def tokenize_sentences(sentence):
|
| 11 |
encoded_dict = tokenizer.encode_plus(
|
| 12 |
sentence,
|
|
|
|
| 1 |
import torch
|
|
|
|
| 2 |
import streamlit as st
|
| 3 |
from transformers import RobertaTokenizer, RobertaForSequenceClassification
|
| 4 |
+
import nltk
|
| 5 |
+
from nltk.corpus import stopwords
|
| 6 |
+
import re
|
| 7 |
nltk.download('stopwords')
|
| 8 |
nltk.download('punkt')
|
| 9 |
stop_words = set(stopwords.words('english'))
|
| 10 |
stop_words.discard('and')
|
| 11 |
|
| 12 |
+
|
| 13 |
def tokenize_sentences(sentence):
|
| 14 |
encoded_dict = tokenizer.encode_plus(
|
| 15 |
sentence,
|