Text Generation
Transformers
Aragonese
Arabic
English
code
phishing
sms
classification
SMS-hpishing-detection
spam
analyzing
art
finance
biology
text-generation-inference
Mixture of Experts
Merge
Instructions to use Noufy/navy-bayes with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Noufy/navy-bayes with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Noufy/navy-bayes")# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Noufy/navy-bayes", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Noufy/navy-bayes with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Noufy/navy-bayes" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Noufy/navy-bayes", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Noufy/navy-bayes
- SGLang
How to use Noufy/navy-bayes with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "Noufy/navy-bayes" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Noufy/navy-bayes", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "Noufy/navy-bayes" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Noufy/navy-bayes", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use Noufy/navy-bayes with Docker Model Runner:
docker model run hf.co/Noufy/navy-bayes
Upload NavyBayes.py
Browse files- NavyBayes.py +50 -61
NavyBayes.py
CHANGED
|
@@ -1,11 +1,11 @@
|
|
| 1 |
-
import firebase_admin
|
| 2 |
-
from firebase_admin import credentials, firestore
|
| 3 |
-
from joblib import dump, load
|
| 4 |
import datetime
|
| 5 |
-
|
| 6 |
-
from sklearn.
|
| 7 |
-
import
|
| 8 |
-
|
| 9 |
|
| 10 |
# التهيئة مرة واحدة فقط
|
| 11 |
if not firebase_admin._apps:
|
|
@@ -25,7 +25,7 @@ except Exception as e:
|
|
| 25 |
vectorizer = None
|
| 26 |
print(f"Model and vectorizer not found. You need to train the model. Error: {e}")
|
| 27 |
|
| 28 |
-
# 1. وظيفة لتحليل و
|
| 29 |
def classify_and_store_message(message):
|
| 30 |
global model, vectorizer
|
| 31 |
try:
|
|
@@ -44,19 +44,13 @@ def classify_and_store_message(message):
|
|
| 44 |
}
|
| 45 |
|
| 46 |
# تخزين الرسالة في مجموعة Firestore حسب التصنيف
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
elif classification == "news_phishing":
|
| 50 |
-
db.collection('news').add(message_data)
|
| 51 |
-
elif classification == "advertisement_phishing":
|
| 52 |
-
db.collection('advertisement').add(message_data)
|
| 53 |
-
elif classification == "social_phishing":
|
| 54 |
-
db.collection('social').add(message_data)
|
| 55 |
|
| 56 |
# تخزين الرسالة في مجموعة 'all_messages' لجميع الرسائل
|
| 57 |
db.collection('all_messages').add(message_data)
|
| 58 |
|
| 59 |
-
# تخزين الرسالة في مجموعة 'recently_analyzed_messages'
|
| 60 |
db.collection('recently_analyzed_messages').add(message_data)
|
| 61 |
|
| 62 |
print(f"Message classified as {classification} and stored in Firestore.")
|
|
@@ -66,38 +60,37 @@ def classify_and_store_message(message):
|
|
| 66 |
print(f"Error classifying message: {e}")
|
| 67 |
return None
|
| 68 |
|
| 69 |
-
# 2.
|
| 70 |
-
def
|
| 71 |
-
|
| 72 |
-
|
| 73 |
-
|
| 74 |
-
|
| 75 |
-
|
| 76 |
-
|
| 77 |
-
|
| 78 |
-
|
| 79 |
-
|
| 80 |
-
|
| 81 |
-
|
| 82 |
-
|
| 83 |
-
|
| 84 |
-
|
| 85 |
-
|
| 86 |
-
|
| 87 |
-
|
| 88 |
-
|
| 89 |
-
|
| 90 |
-
|
| 91 |
-
repo_id="Noufy/https-api_inference.huggingface.comodels-Noufy-naive_bayes_sms",
|
| 92 |
-
repo_type="model"
|
| 93 |
-
)
|
| 94 |
-
|
| 95 |
-
print("Model and vectorizer uploaded successfully to Hugging Face.")
|
| 96 |
|
| 97 |
-
|
| 98 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 99 |
|
| 100 |
-
# 3. تحديث النموذج مع بيانات جديدة
|
| 101 |
def update_model_with_new_data(new_messages, new_labels):
|
| 102 |
global model, vectorizer
|
| 103 |
try:
|
|
@@ -129,36 +122,32 @@ def update_model_with_new_data(new_messages, new_labels):
|
|
| 129 |
except Exception as e:
|
| 130 |
print(f"Error updating model: {e}")
|
| 131 |
|
| 132 |
-
# تشغيل رفع النموذج إلى Hugging Face
|
| 133 |
-
upload_model_to_huggingface()
|
| 134 |
-
|
| 135 |
# 4. دالة لاختبار النظام
|
| 136 |
def test_system():
|
| 137 |
test_messages = [
|
| 138 |
-
"Win a free vacation now!",
|
| 139 |
"Breaking news: Major stock updates today.",
|
| 140 |
"Don't forget our meeting tomorrow at 10 AM.",
|
| 141 |
-
"Click here to secure your bank account
|
| 142 |
-
"Exclusive offers just for you
|
| 143 |
]
|
| 144 |
|
| 145 |
for msg in test_messages:
|
| 146 |
-
|
| 147 |
-
|
| 148 |
|
| 149 |
# 5. وظيفة للتصحيح اليدوي
|
| 150 |
def correct_classification(message_id, correct_label):
|
| 151 |
try:
|
| 152 |
-
# جلب الرسالة من
|
| 153 |
-
message_ref = db.collection('
|
| 154 |
message_data = message_ref.get().to_dict()
|
| 155 |
|
| 156 |
if not message_data:
|
| 157 |
print("Message not found.")
|
| 158 |
return
|
| 159 |
|
| 160 |
-
# تحديث التصنيف في
|
| 161 |
-
message_data['classification'] = correct_label
|
| 162 |
message_ref.update({'classification': correct_label})
|
| 163 |
|
| 164 |
# إضافة البيانات إلى نموذج التدريب الجديد
|
|
@@ -167,5 +156,5 @@ def correct_classification(message_id, correct_label):
|
|
| 167 |
except Exception as e:
|
| 168 |
print(f"Error correcting classification: {e}")
|
| 169 |
|
| 170 |
-
#
|
| 171 |
-
|
|
|
|
| 1 |
+
import firebase_admin # type: ignore
|
| 2 |
+
from firebase_admin import credentials, firestore # type: ignore
|
| 3 |
+
from joblib import dump, load # type: ignore
|
| 4 |
import datetime
|
| 5 |
+
import re
|
| 6 |
+
from sklearn.feature_extraction.text import TfidfVectorizer # type: ignore
|
| 7 |
+
from sklearn.naive_bayes import MultinomialNB # type: ignore
|
| 8 |
+
import pandas as pd # type: ignore
|
| 9 |
|
| 10 |
# التهيئة مرة واحدة فقط
|
| 11 |
if not firebase_admin._apps:
|
|
|
|
| 25 |
vectorizer = None
|
| 26 |
print(f"Model and vectorizer not found. You need to train the model. Error: {e}")
|
| 27 |
|
| 28 |
+
# 1. وظيفة لتحليل النصوص وتصنيفها
|
| 29 |
def classify_and_store_message(message):
|
| 30 |
global model, vectorizer
|
| 31 |
try:
|
|
|
|
| 44 |
}
|
| 45 |
|
| 46 |
# تخزين الرسالة في مجموعة Firestore حسب التصنيف
|
| 47 |
+
collection_name = classification.split('_')[0] # استخدام الجزء الأول من التصنيف كاسم المجموعة
|
| 48 |
+
db.collection(collection_name).add(message_data)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
|
| 50 |
# تخزين الرسالة في مجموعة 'all_messages' لجميع الرسائل
|
| 51 |
db.collection('all_messages').add(message_data)
|
| 52 |
|
| 53 |
+
# تخزين الرسالة في مجموعة 'recently_analyzed_messages'
|
| 54 |
db.collection('recently_analyzed_messages').add(message_data)
|
| 55 |
|
| 56 |
print(f"Message classified as {classification} and stored in Firestore.")
|
|
|
|
| 60 |
print(f"Error classifying message: {e}")
|
| 61 |
return None
|
| 62 |
|
| 63 |
+
# 2. وظيفة لتحليل النصوص المدخلة
|
| 64 |
+
def analyze_input_text():
|
| 65 |
+
print("\n--- SMS Classification and Link Analysis Tool ---")
|
| 66 |
+
while True:
|
| 67 |
+
user_input = input("Enter a message to classify (or type 'exit' to quit): ").strip()
|
| 68 |
+
if user_input.lower() == 'exit':
|
| 69 |
+
print("Exiting the tool. Goodbye!")
|
| 70 |
+
break
|
| 71 |
+
|
| 72 |
+
# استخراج الروابط من النص المدخل
|
| 73 |
+
links = re.findall(r'(https?://[^\s]+)', user_input)
|
| 74 |
+
if links:
|
| 75 |
+
print(f"Detected links: {links}")
|
| 76 |
+
# تحليل الروابط (يمكن تطوير التحليل ليشمل أدوات أو خدمات خارجية)
|
| 77 |
+
for link in links:
|
| 78 |
+
# افتراض تحليل بسيط (يمكن تحسينه لاحقًا)
|
| 79 |
+
if "secure" in link or "safe" in link:
|
| 80 |
+
print(f"Link '{link}' appears safe.")
|
| 81 |
+
else:
|
| 82 |
+
print(f"Link '{link}' might be suspicious.")
|
| 83 |
+
else:
|
| 84 |
+
print("No links detected in the message.")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 85 |
|
| 86 |
+
# تصنيف الرسالة
|
| 87 |
+
classification = classify_and_store_message(user_input)
|
| 88 |
+
if classification:
|
| 89 |
+
print(f"Message classified as: {classification}")
|
| 90 |
+
else:
|
| 91 |
+
print("Unable to classify the message. Please try again.")
|
| 92 |
|
| 93 |
+
# 3. دالة لتحديث النموذج مع بيانات جديدة
|
| 94 |
def update_model_with_new_data(new_messages, new_labels):
|
| 95 |
global model, vectorizer
|
| 96 |
try:
|
|
|
|
| 122 |
except Exception as e:
|
| 123 |
print(f"Error updating model: {e}")
|
| 124 |
|
|
|
|
|
|
|
|
|
|
| 125 |
# 4. دالة لاختبار النظام
|
| 126 |
def test_system():
|
| 127 |
test_messages = [
|
| 128 |
+
"Win a free vacation now! Visit https://spam-link.com",
|
| 129 |
"Breaking news: Major stock updates today.",
|
| 130 |
"Don't forget our meeting tomorrow at 10 AM.",
|
| 131 |
+
"Click here to secure your bank account: https://phishing-link.com",
|
| 132 |
+
"Exclusive offers just for you! Buy now at https://ad-link.com"
|
| 133 |
]
|
| 134 |
|
| 135 |
for msg in test_messages:
|
| 136 |
+
print(f"\nAnalyzing message: {msg}")
|
| 137 |
+
analyze_input_text(msg)
|
| 138 |
|
| 139 |
# 5. وظيفة للتصحيح اليدوي
|
| 140 |
def correct_classification(message_id, correct_label):
|
| 141 |
try:
|
| 142 |
+
# جلب الرسالة من Firestore
|
| 143 |
+
message_ref = db.collection('all_messages').document(message_id)
|
| 144 |
message_data = message_ref.get().to_dict()
|
| 145 |
|
| 146 |
if not message_data:
|
| 147 |
print("Message not found.")
|
| 148 |
return
|
| 149 |
|
| 150 |
+
# تحديث التصنيف في Firestore
|
|
|
|
| 151 |
message_ref.update({'classification': correct_label})
|
| 152 |
|
| 153 |
# إضافة البيانات إلى نموذج التدريب الجديد
|
|
|
|
| 156 |
except Exception as e:
|
| 157 |
print(f"Error correcting classification: {e}")
|
| 158 |
|
| 159 |
+
# تشغيل تحليل النصوص
|
| 160 |
+
analyze_input_text()
|