Merge Develop into Master

.gitignore

@@ -0,0 +1 @@
+.env

.gitlab/merge_request_templates/default.md

@@ -0,0 +1,12 @@
+## 이슈
+
+Jira 이슈 : [S11P31S102-](https://ssafy.atlassian.net/browse/S11P31S102-)
+
+## 개요
+이 Merge Request는 기능을 추가합니다.
+
+## 변경 사항
+- 
+
+## 추가 사항
+- 

.gitmessage.txt

@@ -13,13 +13,14 @@ feat:
 ################
 # 꼬릿말(footer)을 아랫줄에 작성 (현재 커밋과 관련된 이슈 번호 추가 등)
 # 예) Close #7
-티켓: S11P21S102-
+티켓: S11P31S102-
 ################
 # feat : 새로운 기능 추가
 # fix : 버그 수정
 # docs : 문서 수정
 # test : 테스트 코드 추가
-# refact : 코드 리팩토링
+# refactor : 코드 리팩토링
 # style : 코드 의미에 영향을 주지 않는 변경사항
 # chore : 빌드 부분 혹은 패키지 매니저 수정사항
-################
+# design: CSS 등 사용자 UI 디자인 변경
+################

AI/.gitignore

@@ -0,0 +1,206 @@
+# Created by https://www.toptal.com/developers/gitignore/api/visualstudiocode,python
+# Edit at https://www.toptal.com/developers/gitignore?templates=visualstudiocode,python
+
+### PyCharm ###
+.idea
+
+### Python ###
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/#use-with-ide
+.pdm.toml
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/
+
+### Python Patch ###
+# Poetry local configuration file - https://python-poetry.org/docs/configuration/#local-configuration
+poetry.toml
+
+# ruff
+.ruff_cache/
+
+# LSP config files
+pyrightconfig.json
+
+### VisualStudioCode ###
+.vscode/*
+!.vscode/settings.json
+!.vscode/tasks.json
+!.vscode/launch.json
+!.vscode/extensions.json
+!.vscode/*.code-snippets
+
+# Local History for Visual Studio Code
+.history/
+
+# Built Visual Studio Code Extensions
+*.vsix
+
+### VisualStudioCode Patch ###
+# Ignore all local history of files
+.history
+.ionide
+
+# End of https://www.toptal.com/developers/gitignore/api/visualstudiocode,python
+
+### vector db file
+vector_store_path/
+*.faiss
+*.pkl
+
+### local model 
+local_model/

AI/README.md

@@ -0,0 +1,16 @@
+## Example code
+
+### set up file
+
+- chatGPT API를 사용하거나 langsmith에 프로젝트 로그를 남기기 위해서 아래와 같이 .env file을 구성해야 합니다.
+
+- .env 파일을 chain 폴더 안에 위치합니다.
+
+- .env 파일은 아래와 같이 구성합니다.
+```
+OPENAI_API_KEY=YOUR_OPENAI_API_KEY
+LANGCHAIN_TRACING_V2=true
+LANGCHAIN_ENDPOINT=https://api.smith.langchain.com
+LANGCHAIN_API_KEY=YOUR_LANGCHAIN_API_KEY
+LANGCHAIN_PROJECT=YOUR_PROJECT_NAME
+```

AI/sembot/batch.py

@@ -0,0 +1,92 @@
+# batch.py
+
+import torch
+from torch.nn.utils.rnn import pad_sequence
+import asyncio
+from prepare import load_prompt, format_docs, memory_retriever, filter_docs
+from streamer import BatchTextStreamer
+
+def make_generation_kwargs(input_ids, eos_token_id, streamer):
+    generation_kwargs = {
+        "input_ids": input_ids,
+        "max_new_tokens": 400,
+        "eos_token_id": eos_token_id,
+        "do_sample": True,
+        "temperature": 0.1,
+        "top_p": 0.9,
+        "streamer": streamer
+    }
+    return generation_kwargs
+
+
+def generate_model(model, generation_kwargs):
+    with torch.no_grad():
+        model.generate(**generation_kwargs)
+    # 메모리 정리
+    torch.cuda.empty_cache()
+
+
+# 배치 처리 백그라운드 태스크
+async def process_queue(request_queue, batch_size, batch_lock, retriever_lock, event_loop, tokenizer, model, eos_token_id, embeddings, retriever):
+    while True:
+        # 최신 retriever 가져오기
+        async with retriever_lock:
+            current_retriever = retriever
+
+        async with batch_lock:
+            batch = []
+            response_queues = []
+
+
+            # 첫 번째 요청을 기다립니다.
+            try:
+                request = await request_queue.get()
+                input_data = {'question': request['question'], 'memory': request['memory'], 'level': request['level']}
+                batch.append(input_data)
+                response_queues.append(request['response_queue'])
+            except Exception as e:
+                continue  # 예외 처리
+
+            # 추가 요청을 배치 사이즈만큼 수집합니다.
+            while len(batch) < batch_size:
+                try:
+                    # 대기하지 않고 요청이 있으면 즉시 가져옵니다.
+                    request = request_queue.get_nowait()
+                    input_data = {'question': request['question'], 'memory': request['memory'], 'level': request['level']}
+                    batch.append(input_data)
+                    response_queues.append(request['response_queue'])
+                except asyncio.QueueEmpty:
+                    break  # 대기열이 비어 있으면 종료
+
+            if batch:
+                # 배치된 요청들에 대해 메시지 생성
+                messages = []
+                for input_data in batch:
+                    message = []
+                    message.append({"role": "system", "content": load_prompt()})
+                    for qda in input_data["memory"]:
+                        message.append({"role": "user", "content": qda.get("question", "")})
+                        message.append({"role": "assistant", "content": qda.get("answer", "")})
+
+                    # 사용자 레벨에 맞는 문서 필터링
+                    filtered_docs = filter_docs(current_retriever.invoke(input_data['question']), input_data["level"])
+
+                    message.append({"role": "user", "content": f"{format_docs(memory_retriever(input_data, embeddings))}\n\n {format_docs(filtered_docs)}\n\n {input_data['question']}"})
+                    messages.append(message)
+
+                # 배치 입력 생성
+                input_ids = [
+                    tokenizer.apply_chat_template(msg, add_generation_prompt=True, return_tensors="pt").to(model.device).squeeze(0)
+                    for msg in messages
+                ]
+                input_ids = pad_sequence(input_ids, batch_first=True, padding_value=tokenizer.pad_token_id)
+
+                # 커스텀 스트리머 생성
+                streamer = BatchTextStreamer(tokenizer, response_queues, event_loop, eos_token_id=eos_token_id, skip_prompt=True)
+
+                # 모델 배치 추론
+                generation_kwargs = make_generation_kwargs(input_ids, eos_token_id, streamer)
+
+                # 모델 생성을 비동기로 실행하고 완료될 때까지 기다립니다.
+                await asyncio.to_thread(generate_model, model, generation_kwargs)
+                # batch_lock은 여기까지 유지됩니다.

AI/sembot/mongo_client.py

@@ -0,0 +1,37 @@
+from pymongo import MongoClient
+
+def singleton(class_):
+    class class_w(class_):
+      _instance = None
+      _sealed = False
+      def __new__(class_, *args, **kwargs):
+
+        if class_w._instance is None:
+            class_w._instance = super(class_w, class_).__new__(class_)
+            class_w._instance._sealed = False
+        return class_w._instance
+      def __init__(self, *args, **kwargs):
+        if self._sealed:
+          return
+        super(class_w, self).__init__(*args, **kwargs)
+        self._sealed = True
+    class_w.__name__ = class_.__name__
+    return class_w
+
+
+# MongoDB 클라이언트와 데이터베이스 초기화
+@singleton
+class MongoDBClient:
+    def __init__(self, uri="mongodb://root:ssafy@k11s102.p.ssafy.io:27017/?authSource=admin", db_name="sembot", collection_name="regulations"):
+        self.client = MongoClient(uri)
+        self.db_name = db_name
+        self.collection_name = collection_name
+
+    def find_regulations(self):
+        # 모든 규정 데이터를 조회해서 리턴한다.
+        return self.client[self.db_name][self.collection_name].find()
+
+    def close(self):
+        # 애플리케이션 종료 시 호출하여 클라이언트 연결 종료
+        self.client.close()
+        print("MongoDB 클라이언트 연결이 종료되었습니다.")