Spaces:
Build error
Build error
Update app.py
Browse filesstill punkt issue
app.py
CHANGED
|
@@ -14,17 +14,21 @@ from nltk.tokenize import word_tokenize, sent_tokenize
|
|
| 14 |
from sklearn.feature_extraction.text import CountVectorizer
|
| 15 |
|
| 16 |
nltk_data_path = '/home/user/nltk_data'
|
|
|
|
| 17 |
if not os.path.exists(nltk_data_path):
|
| 18 |
os.makedirs(nltk_data_path)
|
|
|
|
|
|
|
|
|
|
| 19 |
nltk.download('punkt', download_dir=nltk_data_path)
|
|
|
|
| 20 |
nltk.download('stopwords', download_dir=nltk_data_path)
|
| 21 |
-
else:
|
| 22 |
-
if not os.path.exists(os.path.join(nltk_data_path, "tokenizers/punkt")):
|
| 23 |
-
nltk.download('punkt', download_dir=nltk_data_path)
|
| 24 |
-
if not os.path.exists(os.path.join(nltk_data_path, "corpora/stopwords")):
|
| 25 |
-
nltk.download('stopwords', download_dir=nltk_data_path)
|
| 26 |
|
| 27 |
-
nltk.
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
|
| 29 |
model = load_model('model_improved.keras')
|
| 30 |
vectorizer = joblib.load('vectorizer.joblib')
|
|
@@ -34,7 +38,6 @@ with open('product_mapping.json', 'r') as file1:
|
|
| 34 |
reverse_mapping = {v: k for k, v in product_mapping.items()}
|
| 35 |
|
| 36 |
lemmatizer = WordNetLemmatizer()
|
| 37 |
-
stop_words = set(stopwords.words('english'))
|
| 38 |
|
| 39 |
def clean_text(text):
|
| 40 |
if text is None:
|
|
|
|
| 14 |
from sklearn.feature_extraction.text import CountVectorizer
|
| 15 |
|
| 16 |
nltk_data_path = '/home/user/nltk_data'
|
| 17 |
+
|
| 18 |
if not os.path.exists(nltk_data_path):
|
| 19 |
os.makedirs(nltk_data_path)
|
| 20 |
+
|
| 21 |
+
nltk.data.path.append(nltk_data_path)
|
| 22 |
+
if not os.path.exists(os.path.join(nltk_data_path, "tokenizers/punkt")):
|
| 23 |
nltk.download('punkt', download_dir=nltk_data_path)
|
| 24 |
+
if not os.path.exists(os.path.join(nltk_data_path, "corpora/stopwords")):
|
| 25 |
nltk.download('stopwords', download_dir=nltk_data_path)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
|
| 27 |
+
from nltk.corpus import stopwords
|
| 28 |
+
from nltk.tokenize import sent_tokenize
|
| 29 |
+
|
| 30 |
+
stop_words = set(stopwords.words('english'))
|
| 31 |
+
|
| 32 |
|
| 33 |
model = load_model('model_improved.keras')
|
| 34 |
vectorizer = joblib.load('vectorizer.joblib')
|
|
|
|
| 38 |
reverse_mapping = {v: k for k, v in product_mapping.items()}
|
| 39 |
|
| 40 |
lemmatizer = WordNetLemmatizer()
|
|
|
|
| 41 |
|
| 42 |
def clean_text(text):
|
| 43 |
if text is None:
|