Initial commit

.gitignore

@@ -0,0 +1,160 @@
+# 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/

LICENSE

@@ -0,0 +1,7 @@
+Copyright <YEAR> <COPYRIGHT HOLDER>
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the “Software”), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

app.py

@@ -0,0 +1,127 @@
+import gradio as gr
+
+from llms.tiny_llama import TinyLlama
+from knowledgebase import KnowledgeBase
+from langchain.chains import ConversationalRetrievalChain
+from langchain.memory import ConversationBufferMemory
+
+
+LLM = None
+KNOWLEDGEBASE = None
+SYTEM_PROMPT = (
+    "You are an assistant. Answer to a user's query based on a given context."
+)
+
+# MEMORY = ConversationBufferMemory(
+#     memory_key="chat_history", output_key="answer", return_messages=True
+# )
+QA_CHAIN = None
+
+
+# def init_qa_chain():
+#     QA_CHAIN = ConversationalRetrievalChain.from_llm(
+#         LLM,
+#         retriever=KNOWLEDGEBASE,
+#         chain_type="stuff",
+#         memory=MEMORY,
+#         return_source_documents=True,
+#         verbose=True,
+#     )
+
+
+def init_llm(llm="TinyLlama"):
+    global LLM
+    LLM = TinyLlama()
+
+
+def chat(message, history):
+    global LLM, KNOWLEDGEBASE, SYTEM_PROMPT
+    context = KNOWLEDGEBASE.invoke(message)[0].page_content
+    system = {"role": "system", "content": SYTEM_PROMPT}
+    user = {
+        "role": "user",
+        "content": f"prompt: ```{message}``\ncontext:```{context}```",
+    }
+    response = LLM([system, user]).split("<|assistant|>")[-1]
+    return response
+
+
+def init_rag(system_prompt, url_input, file_input):
+    global SYTEM_PROMPT
+    if SYTEM_PROMPT != system_prompt:
+        SYTEM_PROMPT = system_prompt
+        gr.Info("Saved new system prompt")
+    if url_input and file_input:
+        gr.Error(message="Provide either an URL or a File")
+    path = url_input if url_input else file_input
+    load_knowledgebase(path)
+
+
+def load_knowledgebase(path):
+    global KNOWLEDGEBASE
+    if not KNOWLEDGEBASE:
+        KNOWLEDGEBASE = KnowledgeBase()
+    print("Loading knowledgebase:", path)
+    if not path:
+        return
+    if "https://" in path:
+        KNOWLEDGEBASE.load_url(path)
+        gr.Info(message="Succesfully loaded URL")
+    else:
+        if path.split(".")[-1] == "pdf":
+            KNOWLEDGEBASE.load_pdf(path)
+            gr.Info(message="Succesfully loaded pdf")
+        else:
+            KNOWLEDGEBASE.load_txt(path)
+            gr.Info(message="Succesfully loaded txt")
+
+
+def show_file(file):
+    print(file)
+    return file
+
+
+with gr.Blocks(title="d-RAG") as iface:
+    gr.Markdown(
+        """# d-RAG &nbsp;[![Watch on GitHub](https://img.shields.io/github/watchers/rumbleFTW/d-RAG.svg?style=social)](https://github.com/rumbleFTW/d-RAG/watchers) &nbsp; [![Star on GitHub](https://img.shields.io/github/stars/rumbleFTW/d-RAG.svg?style=social)](https://github.com/rumbleFTW/d-RAG/stargazers)
+"""
+    )
+    with gr.Row(equal_height=True):
+        with gr.Column():
+            with gr.Row():
+                model = gr.Dropdown(
+                    label="Model",
+                    choices=[
+                        "TinyLlama-1.1B-Chat-v1.0",
+                        "Mixtral-8x7B-Instruct-v0.1",
+                        "Mistral-7B-Instruct-v0.2",
+                    ],
+                    value="TinyLlama-1.1B-Chat-v1.0",
+                    scale=1,
+                    interactive=True,
+                )
+                system_prompt = gr.Textbox(
+                    label="System prompt",
+                    value="You are an assistant. Answer to a user's query based on a given context.",
+                    scale=2,
+                )
+            with gr.Accordion(label="Knowledge base", open=True):
+                url_input = gr.Textbox(placeholder="URL", value=None)
+                gr.Markdown("OR")
+                file_input = gr.File(
+                    file_count="multiple",
+                    file_types=[".txt", ".pdf"],
+                    show_label=True,
+                    visible=True,
+                )
+            submit = gr.Button("Submit")
+            submit.click(
+                fn=init_rag,
+                inputs=[system_prompt, url_input, file_input],
+            )
+        with gr.Column():
+            demo = gr.ChatInterface(fn=chat, examples=["Namaste!", "Hello!", "Hola!"])
+
+if __name__ == "__main__":
+    init_llm()
+    iface.launch(debug=True)

knowledgebase.py

@@ -0,0 +1,43 @@
+from langchain_community.vectorstores.faiss import FAISS
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_community.document_loaders import text, PyPDFLoader, WebBaseLoader
+from langchain_community.document_transformers import Html2TextTransformer
+from langchain_community.embeddings import HuggingFaceEmbeddings
+
+
+class KnowledgeBase:
+    def __init__(self) -> None:
+        self.embeddings = HuggingFaceEmbeddings(
+            model_name="sentence-transformers/all-mpnet-base-v2"
+        )
+        self.text_splitter = RecursiveCharacterTextSplitter(
+            chunk_size=100, chunk_overlap=50
+        )
+        self.retriever = None
+
+    def load_txt(self, path):
+        loader = text.TextLoader(file_path=path)
+        documents = loader.load()
+        chunked_docs = self.text_splitter.split_documents(documents=documents)
+        print(chunked_docs.__len__())
+        db = FAISS.from_documents(embedding=self.embeddings, documents=chunked_docs)
+        self.retriever = db.as_retriever()
+
+    def load_pdf(self, path):
+        loader = PyPDFLoader(file_path=path)
+        documents = loader.load()
+        chunked_docs = self.text_splitter.split_documents(documents=documents)
+        db = FAISS.from_documents(embedding=self.embeddings, documents=chunked_docs)
+        self.retriever = db.as_retriever()
+
+    def load_url(self, path):
+        loader = WebBaseLoader(web_path=path)
+        docs = loader.load()
+        html2text = Html2TextTransformer()
+        docs_transformed = html2text.transform_documents(docs)
+        chunked_docs = self.text_splitter.split_documents(docs_transformed)
+        db = FAISS.from_documents(embedding=self.embeddings, documents=chunked_docs)
+        self.retriever = db.as_retriever()
+
+    def invoke(self, query):
+        return self.retriever.invoke(query)

llms/tiny_llama.py

@@ -0,0 +1,41 @@
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM
+
+
+class TinyLlama:
+    def __init__(self) -> None:
+        self.tokenizer = AutoTokenizer.from_pretrained(
+            "TinyLlama/TinyLlama-1.1B-Chat-v1.0"
+        )
+        self.model = AutoModelForCausalLM.from_pretrained(
+            "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
+            load_in_4bit=True,
+            device_map="auto",
+            bnb_4bit_compute_dtype=torch.float16,
+        )
+
+        print(f"LLM loaded to {self.model.device}")
+
+        self._messages = []
+
+    def __call__(self, messages, *args, **kwds):
+        tokenized_chat = self.tokenizer.apply_chat_template(
+            messages, tokenize=False, add_generation_prompt=True
+        )
+        inputs = self.tokenizer(tokenized_chat, return_tensors="pt").to(
+            self.model.device
+        )
+
+        outputs = self.model.generate(
+            **inputs,
+            use_cache=True,
+            max_length=1000,
+            min_length=10,
+            temperature=0.7,
+            num_return_sequences=1,
+            do_sample=True,
+        )
+
+        generated_text = self.tokenizer.decode(outputs[0], skip_special_tokens=True)
+
+        return generated_text

requirements.txt

@@ -0,0 +1,108 @@
+accelerate==0.26.1
+aiofiles==23.2.1
+aiohttp==3.9.3
+aiosignal==1.3.1
+altair==5.2.0
+annotated-types==0.6.0
+anyio==4.2.0
+attrs==23.2.0
+bitsandbytes==0.42.0
+certifi==2023.11.17
+charset-normalizer==3.3.2
+click==8.1.7
+colorama==0.4.6
+contourpy==1.2.0
+cycler==0.12.1
+dataclasses-json==0.6.3
+fastapi==0.109.0
+ffmpy==0.3.1
+filelock==3.13.1
+fonttools==4.47.2
+frozenlist==1.4.1
+fsspec==2023.12.2
+gradio==4.16.0
+gradio_client==0.8.1
+greenlet==3.0.3
+h11==0.14.0
+httpcore==1.0.2
+httpx==0.26.0
+huggingface-hub==0.20.3
+idna==3.6
+importlib-resources==6.1.1
+Jinja2==3.1.3
+jsonpatch==1.33
+jsonpointer==2.4
+jsonschema==4.21.1
+jsonschema-specifications==2023.12.1
+kiwisolver==1.4.5
+langchain==0.1.4
+langchain-community==0.0.16
+langchain-core==0.1.17
+langsmith==0.0.84
+markdown-it-py==3.0.0
+MarkupSafe==2.1.4
+marshmallow==3.20.2
+matplotlib==3.8.2
+mdurl==0.1.2
+mpmath==1.3.0
+multidict==6.0.4
+mypy-extensions==1.0.0
+networkx==3.2.1
+numpy==1.26.3
+nvidia-cublas-cu12==12.1.3.1
+nvidia-cuda-cupti-cu12==12.1.105
+nvidia-cuda-nvrtc-cu12==12.1.105
+nvidia-cuda-runtime-cu12==12.1.105
+nvidia-cudnn-cu12==8.9.2.26
+nvidia-cufft-cu12==11.0.2.54
+nvidia-curand-cu12==10.3.2.106
+nvidia-cusolver-cu12==11.4.5.107
+nvidia-cusparse-cu12==12.1.0.106
+nvidia-nccl-cu12==2.18.1
+nvidia-nvjitlink-cu12==12.3.101
+nvidia-nvtx-cu12==12.1.105
+orjson==3.9.12
+packaging==23.2
+pandas==2.2.0
+pillow==10.2.0
+psutil==5.9.8
+pydantic==2.6.0
+pydantic_core==2.16.1
+pydub==0.25.1
+Pygments==2.17.2
+pyparsing==3.1.1
+python-dateutil==2.8.2
+python-multipart==0.0.6
+pytz==2023.4
+PyYAML==6.0.1
+referencing==0.33.0
+regex==2023.12.25
+requests==2.31.0
+rich==13.7.0
+rpds-py==0.17.1
+ruff==0.1.15
+safetensors==0.4.2
+scipy==1.12.0
+semantic-version==2.10.0
+shellingham==1.5.4
+six==1.16.0
+sniffio==1.3.0
+SQLAlchemy==2.0.25
+starlette==0.35.1
+sympy==1.12
+tenacity==8.2.3
+tokenizers==0.15.1
+tomlkit==0.12.0
+toolz==0.12.1
+torch==2.1.2
+tqdm==4.66.1
+transformers==4.37.2
+triton==2.1.0
+typer==0.9.0
+typing-inspect==0.9.0
+typing_extensions==4.9.0
+tzdata==2023.4
+urllib3==2.1.0
+uvicorn==0.27.0.post1
+websockets==11.0.3
+yarl==1.9.4