new repo Initial

.devcontainer/Dockerfile

@@ -0,0 +1,15 @@
+ARG VARIANT="3.10-bookworm"
+#FROM mcr.microsoft.com/devcontainers/python:1-${VARIANT}
+FROM langchain/langchain
+
+# [Optional] Uncomment if you want to install an additional version of node using nvm
+# ARG EXTRA_NODE_VERSION=10
+# RUN su node -c "source /usr/local/share/nvm/nvm.sh && nvm install ${EXTRA_NODE_VERSION}"
+
+# [Optional] Uncomment if you want to install more global node modules
+# RUN su node -c "npm install -g <your-package-list-here>"
+
+#COPY library-scripts/github-debian.sh /tmp/library-scripts/
+RUN apt-get update && apt-get update
+RUN pip install --upgrade pip
+#RUN pip install -r requirements.txt

.devcontainer/devcontainer.json

@@ -0,0 +1,37 @@
+// For format details, see https://aka.ms/devcontainer.json. For config options, see the
+// README at: https://github.com/devcontainers/templates/tree/main/src/python
+{
+	"name": "Langchain latest",
+	// Or use a Dockerfile or Docker Compose file. More info: https://containers.dev/guide/dockerfile
+	//"image": "mcr.microsoft.com/devcontainers/python:1-3.10-bookworm",
+
+	// build config for the docker image instead:
+	"build": { "dockerfile": "Dockerfile" },
+
+	// Features to add to the dev container. More info: https://containers.dev/features.
+	// "features": {},
+
+	// Use 'forwardPorts' to make a list of ports inside the container available locally.
+	// "forwardPorts": [],
+
+	// Use 'postCreateCommand' to run commands after the container is created.
+	// "postCreateCommand": "pip3 install --user -r requirements.txt",
+
+	// Configure tool-specific properties.
+	 "customizations": {
+		 // Configure properties specific to VS Code.
+		 "vscode": {
+			//Add the IDs of extensions you want installed when the container is created.
+			"extensions": ["ms-azuretools.vscode-docker"]
+		  }
+	} //,
+
+	// Uncomment to connect as root instead. More info: https://aka.ms/dev-containers-non-root.
+	// "remoteUser": "root"
+
+	// mount the SSH directory on win
+	/*"mounts": [
+		"type=bind,source=${localWorkspaceFolder},target=/work",
+		"type=bind,source=/home/${localEnv:USER}/.ssh,target=/home/vscode/.ssh,readonly"
+	  ]*/
+}

.gitattributes

@@ -0,0 +1,2 @@
+*.sqlite3 filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text

.github/workflows/check_file_size.yaml

@@ -0,0 +1,16 @@
+name: Check file size
+on:               # or directly `on: [push]` to run the action on every push on any branch
+  pull_request:
+    branches: [main]
+
+  # to run this workflow manually from the Actions tab
+  workflow_dispatch:
+
+jobs:
+  sync-to-hub:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Check large files
+        uses: ActionsDesk/lfs-warning@v2.0
+        with:
+          filesizelimit: 10485760 # this is 10MB so we can sync to HF Spaces

.github/workflows/sync_2_hf.yml

@@ -0,0 +1,20 @@
+name: Sync to Hugging Face hub
+on:
+  push:
+    branches: [main]
+
+  # to run this workflow manually from the Actions tab
+  workflow_dispatch:
+
+jobs:
+  sync-to-hub:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+        with:
+          fetch-depth: 0
+          lfs: true
+      - name: Push to hub
+        env:
+          HF_TOKEN: ${{ secrets.HF_TOKEN }}
+        run: git push https://sabazo:$HF_TOKEN@huggingface.co/spaces/sabazo/docs-qachat-demo main

.gitignore

@@ -158,3 +158,11 @@ cython_debug/
 #  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/
+
+# embeddings db
+*.sqlite3*
+*.bin
+*.pickle
+chroma_db/*
+bin
+obj

README.md

@@ -1 +1 @@
-# docs-qachat
+# docu-qachat-demo

app.py

@@ -0,0 +1,137 @@
+# gradio
+import gradio as gr
+import random
+import time
+#boto3 for S3 access
+import boto3
+from botocore import UNSIGNED
+from botocore.client import Config
+# access .env file
+from dotenv import load_dotenv
+#from bs4 import BeautifulSoup
+# HF libraries
+from langchain.llms import HuggingFaceHub
+from langchain.embeddings import HuggingFaceHubEmbeddings
+# vectorestore
+from langchain.vectorstores import Chroma
+from langchain.vectorstores import FAISS
+# retrieval chain
+from langchain.chains import RetrievalQA
+# prompt template
+from langchain.prompts import PromptTemplate
+from langchain.memory import ConversationBufferMemory
+# logging
+#import logging
+import zipfile
+
+# load HF Token
+config = load_dotenv(".env")
+
+
+model_id = HuggingFaceHub(repo_id="HuggingFaceH4/zephyr-7b-beta", model_kwargs={
+    "temperature":0.1, 
+    "max_new_tokens":1024, 
+    "repetition_penalty":1.2, 
+    "streaming": True, 
+    "return_full_text":True
+    })
+
+model_name = "sentence-transformers/multi-qa-mpnet-base-dot-v1"
+embeddings = HuggingFaceHubEmbeddings(repo_id=model_name)
+
+
+s3 = boto3.client('s3', config=Config(signature_version=UNSIGNED))
+
+## Chroma DB
+s3.download_file('rad-rag-demos', 'vectorstores/chroma.sqlite3', './chroma_db/chroma.sqlite3')
+db = Chroma(persist_directory="./chroma_db", embedding_function=embeddings)
+db.get()
+
+## FAISS DB
+# s3.download_file('rad-rag-demos', 'vectorstores/faiss_db_ray.zip', './chroma_db/faiss_db_ray.zip')
+# with zipfile.ZipFile('./chroma_db/faiss_db_ray.zip', 'r') as zip_ref:
+#     zip_ref.extractall('./chroma_db/')
+
+# FAISS_INDEX_PATH='./chroma_db/faiss_db_ray'
+# db = FAISS.load_local(FAISS_INDEX_PATH, embeddings)
+
+retriever = db.as_retriever(search_type = "mmr")#, search_kwargs={'k': 5, 'fetch_k': 25})
+global qa 
+template = """
+You are the friendly documentation buddy Arti, who helps the Human in using RAY, the open-source unified framework for scaling AI and Python applications.\
+    Use the following context (delimited by <ctx></ctx>) and the chat history (delimited by <hs></hs>) to answer the question :
+------
+<ctx>
+{context}
+</ctx>
+------
+<hs>
+{history}
+</hs>
+------
+{question}
+Answer:
+"""
+prompt = PromptTemplate(
+    input_variables=["history", "context", "question"],
+    template=template,
+)
+memory = ConversationBufferMemory(memory_key="history", input_key="question")
+qa = RetrievalQA.from_chain_type(llm=model_id, chain_type="stuff", retriever=retriever, verbose=True, return_source_documents=True, chain_type_kwargs={
+    "verbose": True,
+    "memory": memory,
+    "prompt": prompt
+}
+    )
+
+
+def add_text(history, text):
+    history = history + [(text, None)]
+    return history, ""
+
+def bot(history):
+    response = infer(history[-1][0], history)
+    print(*memory)
+    sources = [doc.metadata.get("source") for doc in response['source_documents']]
+    src_list = '\n'.join(sources)
+    print_this = response['result']+"\n\n\n Sources: \n\n\n"+src_list
+
+    #history[-1][1] = ""
+    #for character in response['result']: #print_this:
+    #    history[-1][1] += character
+    #    time.sleep(0.05)
+    #    yield history
+    history[-1][1] = print_this #response['result']
+    return history
+
+def infer(question, history):
+    query =  question
+    result = qa({"query": query, "history": history, "question": question})
+    return result
+
+css="""
+#col-container {max-width: 700px; margin-left: auto; margin-right: auto;}
+"""
+
+title = """
+<div style="text-align: center;max-width: 700px;">
+    <h1>Chat with your Documentation</h1>
+    <p style="text-align: center;">Chat with Documentation, <br />
+    when everything is ready, you can start asking questions about the docu ;)</p>
+</div>
+"""
+
+with gr.Blocks(css=css) as demo:
+    with gr.Column(elem_id="col-container"):
+        gr.HTML(title)      
+        chatbot = gr.Chatbot([], elem_id="chatbot")
+        clear = gr.Button("Clear")
+        with gr.Row():
+            question = gr.Textbox(label="Question", placeholder="Type your question and hit Enter ")
+    question.submit(add_text, [chatbot, question], [chatbot, question], queue=False).then(
+        bot, chatbot, chatbot
+    )
+    clear.click(lambda: None, None, chatbot, queue=False)
+
+demo.queue()
+demo.launch()

example.env

@@ -0,0 +1 @@
+OPENAI_API_KEY=

localembedding.py

@@ -0,0 +1,37 @@
+import chromadb
+from chromadb.config import Settings
+from langchain.vectorstores import Chroma
+from langchain.vectorstores.utils import filter_complex_metadata
+import time
+from langchain.embeddings import HuggingFaceEmbeddings
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+
+text_splitter = RecursiveCharacterTextSplitter(
+    # Set a really small chunk size, just to show.
+    chunk_size = 1000,
+    chunk_overlap  = 100,
+    length_function = len,
+)
+
+# Stage one: read all the docs, split them into chunks.
+st = time.time()
+print('Loading documents ...')
+docs = loader.load()
+chunks = text_splitter.create_documents([doc.page_content for doc in docs], metadatas=[doc.metadata for doc in docs])
+et = time.time() - st
+print(f'Time taken: {et} seconds.')
+
+#Stage two: embed the docs.
+# use all-mpnet-base-v2 sentence transformer to convert pieces of text in vectors to store them in the vector store
+model_name = "sentence-transformers/all-mpnet-base-v2"
+model_kwargs = {"device": "cuda"}
+
+embeddings = HuggingFaceEmbeddings(
+    model_name=model_name,
+    model_kwargs=model_kwargs
+    )
+print(f'Loading chunks into vector store ...')
+st = time.time()
+db = Chroma.from_documents(filter_complex_metadata(chunks), embeddings, persist_directory="/content/chroma_db")
+et = time.time() - st
+print(f'Time taken: {et} seconds.')

requirements.txt

@@ -0,0 +1,8 @@
+huggingface_hub
+chromadb
+faiss-cpu
+langchain
+python-dotenv
+bs4
+gradio
+boto3