Spaces:
Running
Running
Commit
·
5bea1cc
1
Parent(s):
d7ef155
Update requirements.txt dan Tambahkan handler NLTK di preprocessing.py
Browse files- preprocessing.py +12 -0
- requirements.txt +2 -2
preprocessing.py
CHANGED
|
@@ -7,6 +7,18 @@ Modul untuk preprocessing teks sebelum prediksi ABSA
|
|
| 7 |
import re
|
| 8 |
import string
|
| 9 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
# Stopwords Indonesia
|
| 11 |
INDONESIAN_STOPWORDS = set([
|
| 12 |
'ada', 'adalah', 'adanya', 'adapun', 'agak', 'agaknya', 'agar', 'akan', 'akankah', 'akhir',
|
|
|
|
| 7 |
import re
|
| 8 |
import string
|
| 9 |
|
| 10 |
+
# ✅ TAMBAHKAN INI - Download NLTK data jika diperlukan
|
| 11 |
+
try:
|
| 12 |
+
import nltk
|
| 13 |
+
try:
|
| 14 |
+
nltk.data.find('tokenizers/punkt_tab')
|
| 15 |
+
except LookupError:
|
| 16 |
+
print("📥 Downloading NLTK punkt_tab...")
|
| 17 |
+
nltk.download('punkt_tab', quiet=True)
|
| 18 |
+
print("✅ NLTK punkt_tab downloaded")
|
| 19 |
+
except ImportError:
|
| 20 |
+
print("⚠️ NLTK tidak terinstall, menggunakan tokenizer sederhana")
|
| 21 |
+
|
| 22 |
# Stopwords Indonesia
|
| 23 |
INDONESIAN_STOPWORDS = set([
|
| 24 |
'ada', 'adalah', 'adanya', 'adapun', 'agak', 'agaknya', 'agar', 'akan', 'akankah', 'akhir',
|
requirements.txt
CHANGED
|
@@ -3,10 +3,10 @@ joblib==1.5.2
|
|
| 3 |
packaging==25.0
|
| 4 |
pandas==2.3.3
|
| 5 |
plotly==6.3.0
|
| 6 |
-
Sastrawi==1.0.1
|
| 7 |
streamlit==1.50.0
|
| 8 |
torch==2.8.0
|
| 9 |
transformers==4.56.2
|
| 10 |
scikit-learn
|
| 11 |
openpyxl
|
| 12 |
-
plotly
|
|
|
|
|
|
| 3 |
packaging==25.0
|
| 4 |
pandas==2.3.3
|
| 5 |
plotly==6.3.0
|
|
|
|
| 6 |
streamlit==1.50.0
|
| 7 |
torch==2.8.0
|
| 8 |
transformers==4.56.2
|
| 9 |
scikit-learn
|
| 10 |
openpyxl
|
| 11 |
+
plotly
|
| 12 |
+
nltk
|