Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -6,23 +6,23 @@ import re
|
|
| 6 |
import os
|
| 7 |
from typing import List, Tuple
|
| 8 |
|
| 9 |
-
# إعداد تسجيل
|
| 10 |
logging.basicConfig(level=logging.INFO)
|
| 11 |
logger = logging.getLogger(__name__)
|
| 12 |
|
| 13 |
-
# تحميل نموذج
|
| 14 |
try:
|
| 15 |
model = SentenceTransformer("UBC-NLP/MARBERT", device="cuda" if torch.cuda.is_available() else "cpu")
|
| 16 |
logger.info("تم تحميل النموذج بنجاح")
|
| 17 |
except Exception as e:
|
| 18 |
-
logger.error(f"
|
| 19 |
raise
|
| 20 |
|
| 21 |
-
# تحميل المعرفة
|
| 22 |
def load_knowledge() -> List[str]:
|
| 23 |
if not os.path.exists("knowledge.txt"):
|
| 24 |
logger.warning("ملف المعرفة غير موجود.")
|
| 25 |
-
return ["عام: ل
|
| 26 |
|
| 27 |
chunks = []
|
| 28 |
current_section = "عام"
|
|
@@ -34,16 +34,16 @@ def load_knowledge() -> List[str]:
|
|
| 34 |
elif line:
|
| 35 |
chunks.append(f"{current_section}: {line}")
|
| 36 |
|
| 37 |
-
logger.info(f"تم تحميل
|
| 38 |
return chunks
|
| 39 |
|
| 40 |
knowledge_chunks = load_knowledge()
|
| 41 |
knowledge_embeddings = model.encode(knowledge_chunks, convert_to_tensor=True)
|
| 42 |
|
| 43 |
-
#
|
| 44 |
def preprocess_question(question: str) -> str:
|
| 45 |
question = re.sub(r'[؟\?،,\.]', '', question).strip()
|
| 46 |
-
|
| 47 |
r'\bماهي\b': 'ما هي',
|
| 48 |
r'\bماهو\b': 'ما هو',
|
| 49 |
r'\bكيفية\b': 'كيف',
|
|
@@ -52,15 +52,15 @@ def preprocess_question(question: str) -> str:
|
|
| 52 |
r'\bعاوز\b': 'كيف يمكن',
|
| 53 |
r'\bعايز\b': 'كيف يمكن'
|
| 54 |
}
|
| 55 |
-
for pattern,
|
| 56 |
-
question = re.sub(pattern,
|
| 57 |
return question
|
| 58 |
|
| 59 |
-
# توليد الرد
|
| 60 |
def generate_response(question: str, top_chunks: List[Tuple[str, float]]) -> str:
|
| 61 |
if not top_chunks:
|
| 62 |
suggestions = [
|
| 63 |
-
"أعد صياغة سؤالك
|
| 64 |
"جرّب استخدام كلمات أخرى.",
|
| 65 |
"ابحث في قسم مختلف من المعرفة."
|
| 66 |
]
|
|
@@ -76,15 +76,13 @@ def generate_response(question: str, top_chunks: List[Tuple[str, float]]) -> str
|
|
| 76 |
sections[section] = []
|
| 77 |
sections[section].append((content.strip(), score))
|
| 78 |
|
| 79 |
-
|
| 80 |
-
main_section = max(sections.keys(), key=lambda k: sum(s for _, s in sections[k]) / len(sections[k]))
|
| 81 |
|
| 82 |
response = f"سؤالك: {question}\n\n"
|
| 83 |
response += f"{main_section}:\n"
|
| 84 |
for content, _ in sorted(sections[main_section], key=lambda x: x[1], reverse=True):
|
| 85 |
response += f"- {content}\n"
|
| 86 |
|
| 87 |
-
# معلومات إضافية
|
| 88 |
other_sections = [s for s in sections if s != main_section]
|
| 89 |
if other_sections:
|
| 90 |
response += "\nمعلومات إضافية:\n"
|
|
@@ -92,16 +90,16 @@ def generate_response(question: str, top_chunks: List[Tuple[str, float]]) -> str
|
|
| 92 |
response += f"\nمن {section}:\n"
|
| 93 |
for content, _ in sorted(sections[section], key=lambda x: x[1], reverse=True)[:2]:
|
| 94 |
response += f"- {content}\n"
|
| 95 |
-
|
| 96 |
return response
|
| 97 |
|
| 98 |
-
#
|
| 99 |
def answer_question(question: str) -> str:
|
| 100 |
if not question or len(question.strip()) < 3:
|
| 101 |
return "يرجى إدخال سؤال واضح لا يقل عن ثلاث كلمات."
|
| 102 |
|
| 103 |
question = preprocess_question(question)
|
| 104 |
-
logger.info(f"
|
| 105 |
|
| 106 |
try:
|
| 107 |
q_embedding = model.encode(question, convert_to_tensor=True)
|
|
@@ -122,25 +120,25 @@ def answer_question(question: str) -> str:
|
|
| 122 |
return generate_response(question, top_chunks)
|
| 123 |
|
| 124 |
except Exception as e:
|
| 125 |
-
logger.error(f"
|
| 126 |
-
return "حدث خطأ أثناء توليد الإجابة.
|
| 127 |
|
| 128 |
# واجهة Gradio
|
| 129 |
-
with gr.Blocks(css=
|
| 130 |
-
with gr.Column(elem_classes=
|
| 131 |
-
gr.Markdown(
|
| 132 |
-
question = gr.Textbox(label=
|
| 133 |
-
submit_btn = gr.Button(
|
| 134 |
-
answer = gr.Textbox(label=
|
| 135 |
gr.Examples(
|
| 136 |
examples=[
|
| 137 |
-
[
|
| 138 |
-
[
|
| 139 |
-
[
|
| 140 |
],
|
| 141 |
inputs=question
|
| 142 |
)
|
| 143 |
submit_btn.click(answer_question, inputs=question, outputs=answer)
|
| 144 |
|
| 145 |
-
if __name__ ==
|
| 146 |
-
demo.launch(server_name=
|
|
|
|
| 6 |
import os
|
| 7 |
from typing import List, Tuple
|
| 8 |
|
| 9 |
+
# إعداد التسجيل
|
| 10 |
logging.basicConfig(level=logging.INFO)
|
| 11 |
logger = logging.getLogger(__name__)
|
| 12 |
|
| 13 |
+
# تحميل نموذج MARBERT
|
| 14 |
try:
|
| 15 |
model = SentenceTransformer("UBC-NLP/MARBERT", device="cuda" if torch.cuda.is_available() else "cpu")
|
| 16 |
logger.info("تم تحميل النموذج بنجاح")
|
| 17 |
except Exception as e:
|
| 18 |
+
logger.error(f"فشل تحميل النموذج: {e}")
|
| 19 |
raise
|
| 20 |
|
| 21 |
+
# تحميل المعرفة
|
| 22 |
def load_knowledge() -> List[str]:
|
| 23 |
if not os.path.exists("knowledge.txt"):
|
| 24 |
logger.warning("ملف المعرفة غير موجود.")
|
| 25 |
+
return ["عام: لا يوجد محتوى معرفي متاح."]
|
| 26 |
|
| 27 |
chunks = []
|
| 28 |
current_section = "عام"
|
|
|
|
| 34 |
elif line:
|
| 35 |
chunks.append(f"{current_section}: {line}")
|
| 36 |
|
| 37 |
+
logger.info(f"تم تحميل {len(chunks)} قطعة معرفة.")
|
| 38 |
return chunks
|
| 39 |
|
| 40 |
knowledge_chunks = load_knowledge()
|
| 41 |
knowledge_embeddings = model.encode(knowledge_chunks, convert_to_tensor=True)
|
| 42 |
|
| 43 |
+
# معالجة السؤال
|
| 44 |
def preprocess_question(question: str) -> str:
|
| 45 |
question = re.sub(r'[؟\?،,\.]', '', question).strip()
|
| 46 |
+
replacements = {
|
| 47 |
r'\bماهي\b': 'ما هي',
|
| 48 |
r'\bماهو\b': 'ما هو',
|
| 49 |
r'\bكيفية\b': 'كيف',
|
|
|
|
| 52 |
r'\bعاوز\b': 'كيف يمكن',
|
| 53 |
r'\bعايز\b': 'كيف يمكن'
|
| 54 |
}
|
| 55 |
+
for pattern, repl in replacements.items():
|
| 56 |
+
question = re.sub(pattern, repl, question)
|
| 57 |
return question
|
| 58 |
|
| 59 |
+
# توليد الرد
|
| 60 |
def generate_response(question: str, top_chunks: List[Tuple[str, float]]) -> str:
|
| 61 |
if not top_chunks:
|
| 62 |
suggestions = [
|
| 63 |
+
"أعد صياغة سؤالك.",
|
| 64 |
"جرّب استخدام كلمات أخرى.",
|
| 65 |
"ابحث في قسم مختلف من المعرفة."
|
| 66 |
]
|
|
|
|
| 76 |
sections[section] = []
|
| 77 |
sections[section].append((content.strip(), score))
|
| 78 |
|
| 79 |
+
main_section = max(sections, key=lambda s: sum(x[1] for x in sections[s]) / len(sections[s]))
|
|
|
|
| 80 |
|
| 81 |
response = f"سؤالك: {question}\n\n"
|
| 82 |
response += f"{main_section}:\n"
|
| 83 |
for content, _ in sorted(sections[main_section], key=lambda x: x[1], reverse=True):
|
| 84 |
response += f"- {content}\n"
|
| 85 |
|
|
|
|
| 86 |
other_sections = [s for s in sections if s != main_section]
|
| 87 |
if other_sections:
|
| 88 |
response += "\nمعلومات إضافية:\n"
|
|
|
|
| 90 |
response += f"\nمن {section}:\n"
|
| 91 |
for content, _ in sorted(sections[section], key=lambda x: x[1], reverse=True)[:2]:
|
| 92 |
response += f"- {content}\n"
|
| 93 |
+
|
| 94 |
return response
|
| 95 |
|
| 96 |
+
# الرد النهائي
|
| 97 |
def answer_question(question: str) -> str:
|
| 98 |
if not question or len(question.strip()) < 3:
|
| 99 |
return "يرجى إدخال سؤال واضح لا يقل عن ثلاث كلمات."
|
| 100 |
|
| 101 |
question = preprocess_question(question)
|
| 102 |
+
logger.info(f"معالجة السؤال: {question}")
|
| 103 |
|
| 104 |
try:
|
| 105 |
q_embedding = model.encode(question, convert_to_tensor=True)
|
|
|
|
| 120 |
return generate_response(question, top_chunks)
|
| 121 |
|
| 122 |
except Exception as e:
|
| 123 |
+
logger.error(f"خطأ أثناء توليد الإجابة: {e}")
|
| 124 |
+
return "حدث خطأ غير متوقع أثناء توليد الإجابة."
|
| 125 |
|
| 126 |
# واجهة Gradio
|
| 127 |
+
with gr.Blocks(css=".arabic-ui {direction: rtl; text-align: right; font-family: Tahoma;}") as demo:
|
| 128 |
+
with gr.Column(elem_classes="arabic-ui"):
|
| 129 |
+
gr.Markdown("### المساعد الذكي لوحدة الشفافية\nاسأل عن الموازنة، المشاركة المجتمعية، أو الشفافية المالية.")
|
| 130 |
+
question = gr.Textbox(label="سؤالك", placeholder="مثال: ما هي أهداف الموازنة التشاركية؟", lines=3)
|
| 131 |
+
submit_btn = gr.Button("إرسال السؤال", variant="primary")
|
| 132 |
+
answer = gr.Textbox(label="الإجابة", lines=12, interactive=False)
|
| 133 |
gr.Examples(
|
| 134 |
examples=[
|
| 135 |
+
["ما هي أهداف الموازنة التشاركية؟"],
|
| 136 |
+
["من هي رئيسة وحدة الشفافية؟"],
|
| 137 |
+
["ما هو دور المواطن في إعداد الموازنة؟"]
|
| 138 |
],
|
| 139 |
inputs=question
|
| 140 |
)
|
| 141 |
submit_btn.click(answer_question, inputs=question, outputs=answer)
|
| 142 |
|
| 143 |
+
if __name__ == "__main__":
|
| 144 |
+
demo.launch(server_name="0.0.0.0", server_port=7860, share=False)
|