chore: Add langchain_faiss to git-lfs tracking

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+langchain_faiss/* filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,161 @@
+# 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/
+

app.py

@@ -1,63 +1,232 @@
+import os
+
 import gradio as gr
-from huggingface_hub import InferenceClient
+from dotenv import load_dotenv
+from langchain.callbacks.base import BaseCallbackHandler
+from langchain.embeddings import CacheBackedEmbeddings
+from langchain.retrievers import BM25Retriever, EnsembleRetriever
+from langchain.storage import LocalFileStore
+from langchain_anthropic import ChatAnthropic
+from langchain_community.chat_models import ChatOllama
+from langchain_community.document_loaders import NotebookLoader, TextLoader
+from langchain_community.document_loaders.generic import GenericLoader
+from langchain_community.document_loaders.parsers.language.language_parser import (
+    LanguageParser,
+)
+from langchain_community.embeddings import HuggingFaceBgeEmbeddings
+from langchain_community.vectorstores import FAISS
+from langchain_core.callbacks.manager import CallbackManager
+from langchain_core.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
+from langchain_core.output_parsers import StrOutputParser
+from langchain_core.prompts import PromptTemplate
+from langchain_core.runnables import ConfigurableField, RunnablePassthrough
+from langchain_google_genai import GoogleGenerativeAI
+from langchain_groq import ChatGroq
+from langchain_openai import ChatOpenAI, OpenAIEmbeddings
+from langchain_text_splitters import Language, RecursiveCharacterTextSplitter
 
+# Load environment variables
+load_dotenv()
+
+# Repository directories
+repo_root_dir = "./docs/langchain"
+repo_dirs = [
+    "libs/core/langchain_core",
+    "libs/community/langchain_community",
+    "libs/experimental/langchain_experimental",
+    "libs/partners",
+    "libs/cookbook",
+]
+repo_dirs = [os.path.join(repo_root_dir, repo) for repo in repo_dirs]
+
+# Load Python documents
+py_documents = []
+for path in repo_dirs:
+    py_loader = GenericLoader.from_filesystem(
+        path,
+        glob="**/*",
+        suffixes=[".py"],
+        parser=LanguageParser(language=Language.PYTHON, parser_threshold=30),
+    )
+    py_documents.extend(py_loader.load())
+print(f".py 파일의 개수: {len(py_documents)}")
+
+# Load Markdown documents
+mdx_documents = []
+for dirpath, _, filenames in os.walk(repo_root_dir):
+    for file in filenames:
+        if file.endswith(".mdx") and "*venv/" not in dirpath:
+            try:
+                mdx_loader = TextLoader(os.path.join(dirpath, file), encoding="utf-8")
+                mdx_documents.extend(mdx_loader.load())
+            except Exception:
+                pass
+print(f".mdx 파일의 개수: {len(mdx_documents)}")
+
+# Load Jupyter Notebook documents
+ipynb_documents = []
+for dirpath, _, filenames in os.walk(repo_root_dir):
+    for file in filenames:
+        if file.endswith(".ipynb") and "*venv/" not in dirpath:
+            try:
+                ipynb_loader = NotebookLoader(
+                    os.path.join(dirpath, file),
+                    include_outputs=True,
+                    max_output_length=20,
+                    remove_newline=True,
+                )
+                ipynb_documents.extend(ipynb_loader.load())
+            except Exception:
+                pass
+print(f".ipynb 파일의 개수: {len(ipynb_documents)}")
+
+
+# Split documents into chunks
+def split_documents(documents, language, chunk_size=2000, chunk_overlap=200):
+    splitter = RecursiveCharacterTextSplitter.from_language(
+        language=language, chunk_size=chunk_size, chunk_overlap=chunk_overlap
+    )
+    return splitter.split_documents(documents)
+
+
+py_docs = split_documents(py_documents, Language.PYTHON)
+mdx_docs = split_documents(mdx_documents, Language.MARKDOWN)
+ipynb_docs = split_documents(ipynb_documents, Language.PYTHON)
+
+print(f"분할된 .py 파일의 개수: {len(py_docs)}")
+print(f"분할된 .mdx 파일의 개수: {len(mdx_docs)}")
+print(f"분할된 .ipynb 파일의 개수: {len(ipynb_docs)}")
+
+combined_documents = py_docs + mdx_docs + ipynb_docs
+print(f"총 도큐먼트 개수: {len(combined_documents)}")
+
+# Initialize embeddings and cache
+store = LocalFileStore("~/.cache/embedding")
+embeddings = HuggingFaceBgeEmbeddings(
+    model_name="BAAI/bge-m3",
+    model_kwargs={"device": "mps"},
+    encode_kwargs={"normalize_embeddings": True},
+)
+cached_embeddings = CacheBackedEmbeddings.from_bytes_store(
+    embeddings, store, namespace=embeddings.model_name
+)
+
+# Create and save FAISS index
+FAISS_DB_INDEX = "./langchain_faiss"
+# db = FAISS.from_documents(combined_documents, cached_embeddings)
+# db.save_local(folder_path=FAISS_DB_INDEX)
+db = FAISS.load_local(
+    FAISS_DB_INDEX, cached_embeddings, allow_dangerous_deserialization=True
+)
+
+# Create retrievers
+faiss_retriever = db.as_retriever(search_type="mmr", search_kwargs={"k": 10})
+bm25_retriever = BM25Retriever.from_documents(combined_documents)
+bm25_retriever.k = 10
+ensemble_retriever = EnsembleRetriever(
+    retrievers=[bm25_retriever, faiss_retriever], weights=[0.5, 0.5], search_type="mmr"
+)
+
+# Create prompt template
+prompt = PromptTemplate.from_template(
+    """당신은 20년차 AI 개발자입니다. 당신의 임무는 주어진 질문에 대하여 최대한 문서의 정보를 활용하여 답변하는 것입니다.
+문서는 Python 코드에 대한 정보를 담고 있습니다. 따라서, 답변을 작성할 때에는 Python 코드에 대한 상세한 code snippet을 포함하여 작성해주세요.
+최대한 자세하게 답변하고, 한글로 답변해 주세요. 주어진 문서에서 답변을 찾을 수 없는 경우, "문서에 답변이 없습니다."라고 답변해 주세요.
+답변은 출처(source)를 반드시 표기해 주세요.
+
+#참고문서:
+{context}
+
+#질문:
+{question}
+
+#답변: 
+
+출처:
+- source1
+- source2
+- ...
 """
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+)
 
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+# Define callback handler for streaming
+class StreamCallback(BaseCallbackHandler):
+    def on_llm_new_token(self, token: str, **kwargs):
+        print(token, end="", flush=True)
+
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+# Initialize LLMs with configuration
+llm = ChatOpenAI(
+    model="gpt-4o",
+    temperature=0,
+    streaming=True,
+    callbacks=[StreamCallback()],
+).configurable_alternatives(
+    ConfigurableField(id="llm"),
+    default_key="gpt4",
+    claude=ChatAnthropic(
+        model="claude-3-opus-20240229",
+        temperature=0,
+        streaming=True,
+        callbacks=[StreamCallback()],
+    ),
+    gpt3=ChatOpenAI(
+        model="gpt-3.5-turbo",
+        temperature=0,
+        streaming=True,
+        callbacks=[StreamCallback()],
+    ),
+    gemini=GoogleGenerativeAI(
+        model="gemini-1.5-flash",
+        temperature=0,
+        streaming=True,
+        callbacks=[StreamCallback()],
+    ),
+    llama3=ChatGroq(
+        model_name="llama3-70b-8192",
+        temperature=0,
+        streaming=True,
+        callbacks=[StreamCallback()],
+    ),
+    ollama=ChatOllama(
+        model="EEVE-Korean-10.8B:long",
+        callback_manager=CallbackManager([StreamingStdOutCallbackHandler()]),
+    ),
+)
 
-    messages.append({"role": "user", "content": message})
+# Create retrieval-augmented generation chain
+rag_chain = (
+    {"context": ensemble_retriever, "question": RunnablePassthrough()}
+    | prompt
+    | llm
+    | StrOutputParser()
+)
 
-    response = ""
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+model_key = os.getenv("LLM_MODEL", "gpt4")
+print("model", model_key)
 
-        response += token
+
+def respond(
+    message,
+    history: list[tuple[str, str]],
+):
+    response = ""
+    for chunk in rag_chain.with_config(configurable={"llm": model_key}).stream(message):
+        response += chunk
         yield response
 
+
 """
 For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
 """
 demo = gr.ChatInterface(
     respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
+    title="랭체인에 대해서 물어보세요!",
+    description="안녕하세요!\n저는 랭체인에 대한 인공지능 QA봇입니다. 랭체인에 대해 깊은 지식을 가지고 있어요. 랭체인 개발에 관한 도움이 필요하시면 언제든지 질문해주세요!",
 )
 
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()

langchain_faiss/index.faiss

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f50b9cdc2968dd1fe5875e7e1f8ed2689a3e938d505a0e2f06b5257083339bd2
+size 2621485

langchain_faiss/index.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba84841f2d61493243e47d654cff88c8864c5fa6119469b7569677e4f82f3c5f
+size 862597

requirements.txt

@@ -1 +1,12 @@
-huggingface_hub==0.22.2
+huggingface_hub==0.22.2
+faiss-cpu
+transformers
+python-dotenv
+langchain
+langchain-anthropic
+langchain-community
+langchain-core
+langchain-google-genai
+langchain-groq
+langchain-openai
+langchain-text-splitters