Upload folder using huggingface_hub

.github/workflows/update_space.yml

@@ -0,0 +1,28 @@
+name: Run Python script
+
+on:
+  push:
+    branches:
+      - main
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+
+    steps:
+    - name: Checkout
+      uses: actions/checkout@v2
+
+    - name: Set up Python
+      uses: actions/setup-python@v2
+      with:
+        python-version: '3.9'
+
+    - name: Install Gradio
+      run: python -m pip install gradio
+
+    - name: Log in to Hugging Face
+      run: python -c 'import huggingface_hub; huggingface_hub.login(token="${{ secrets.hf_token }}")'
+
+    - name: Deploy to Spaces
+      run: gradio deploy

.gitignore

@@ -0,0 +1,11 @@
+# Python-generated files
+__pycache__/
+*.py[oc]
+build/
+dist/
+wheels/
+*.egg-info
+.env
+
+# Virtual environments
+.venv

.python-version

@@ -0,0 +1 @@
+3.12

README.md

@@ -1,12 +1,6 @@
 ---
-title: Bhagavat Gita Chat
-emoji: 😻
-colorFrom: pink
-colorTo: yellow
+title: Bhagavat_Gita_Chat
+app_file: main.py
 sdk: gradio
 sdk_version: 5.38.0
-app_file: app.py
-pinned: false
 ---
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

db.py

@@ -0,0 +1,78 @@
+import chromadb
+from chromadb.config import Settings
+import json
+import csv
+from langchain_community.document_loaders import PyPDFLoader
+from langchain.text_splitter import CharacterTextSplitter
+from langchain_community.embeddings import OpenAIEmbeddings
+
+
+class MyDatabase:
+    def __init__(self):
+        # Settings(persist_directory="./chroma_db")
+        self.chroma_client = chromadb.Client()
+        self.initialize()
+
+    def get_collection(self):
+        return self.chroma_client.get_or_create_collection(name="bhagavat_gita")
+
+    def initialize(self):
+        print("Adding Data ...")
+        collection = self.get_collection()
+        # Read CSV data into a list of dictionaries
+        print("Loading Bhagavat Gita ...")
+        with open(
+            "./data/gita_data.csv", mode="r", newline="", encoding="utf-8"
+        ) as csvfile:
+            documents = list(csv.DictReader(csvfile))
+            # with open("./gita_data.json", "r") as f:
+            # documents = json.load(f)
+            with open("./data/gita_data_new.json", "w") as f:
+                json.dump(documents, f, indent=1)
+            collection.add(
+                documents=[document["translation"] for document in documents],
+                metadatas=[
+                    {
+                        "source": "bhagavat_gita",
+                        "chapter_number": document["chapter_number"],
+                        "verse_number": document["chapter_verse"],
+                    }
+                    for document in documents
+                ],
+                # [
+                #     {"source": "article1"},
+                #     {"source": "article2"},
+                #     {"source": "article3"},
+                # ],
+                # ids=["doc1", "doc2", "doc3"],
+                ids=[f"doc{i}" for i, document in enumerate(documents)],
+            )
+
+        # print("Loading Vishnu Puranam ...")
+        # loader = PyPDFLoader("./data/vishnu_puranam.pdf")
+        # pdfDocument = loader.load()
+        # print("pdfDocument", pdfDocument)
+        # with open("./data/vishnu_puranam.json","w") as f:
+        #         json.dump([doc.model_dump_json() for doc in pdfDocument], f, indent=1)
+
+        # text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=10)
+        # chunked_documents = text_splitter.split_documents([pdfDocument])
+        # print(chunked_documents)
+        print("Added data ...")
+
+    def get_data(self, query: str = "is knowledge superior to action?"):
+        print("Querying data ...")
+        collection = self.get_collection()
+        results = collection.query(
+            query_texts=[
+                query,
+            ],  # Chroma will embed this for you
+            n_results=5,  # how many results to return
+        )
+        print(json.dumps(results, indent=2))
+        return results
+
+
+# mydb = MyDatabase()
+# mydb.initialize()
+# mydb.get_data("What is karma?")

main.py

@@ -0,0 +1,94 @@
+from typing import TypedDict, override
+from langgraph.constants import END, START
+from langgraph.graph.state import StateGraph
+from typing_extensions import Annotated
+from pydantic import BaseModel
+from langgraph.graph.message import add_messages
+import gradio as gr
+from langchain_openai import ChatOpenAI
+from dotenv import load_dotenv
+
+from db import MyDatabase
+
+load_dotenv(override=True)
+mydb = MyDatabase()
+
+
+class State(TypedDict):
+    messages: Annotated[list, add_messages]
+
+
+graph_builder = StateGraph(State)
+
+llm = ChatOpenAI(model="gpt-4o-mini")
+
+
+def chatNode(state: State):
+    messages = state["messages"]
+    print("messages = ", messages)
+    responseMessage = llm.invoke(messages)
+    newState = State(messages=[responseMessage])
+    return newState
+
+
+def encryptNode(state: State):
+    messages = state["messages"]
+    messages[-1].content += "\n--------- \n with love, \n##### Krishna"
+    newState = State(messages=messages)
+    return newState
+
+
+graph_builder.add_node("MyChatNode", chatNode)
+graph_builder.add_node("MyEncryptNode", encryptNode)
+graph_builder.add_edge(START, "MyChatNode")
+graph_builder.add_edge("MyChatNode", "MyEncryptNode")
+graph_builder.add_edge("MyEncryptNode", END)
+graph = graph_builder.compile()
+
+
+def chat(message, history):
+    # Ensure history is a list of message dicts
+    relevant_sections = mydb.get_data(message)
+    if not history:
+        history = [
+            {
+                "role": "system",
+                "content": f"""You are a religious researcher, expert in Hindu literature like Bhagavat Gita. 
+                User asks questions and you will answer from the context given below. it is important that you answer ONLY from the context given below and nowhere else.
+                In your response, mention which chapter and verses from which you came up with this explanation.
+                DO NOT talk about other spiritual traditions. Limit yourself to the context at all times.
+                organize your response under subheadings for clarity and keep it simple in terms of language and brief. Do not add your interpretation or additional commentary.
+                Answer any question in the context of Bhagavat Gita (particularly from the context given below). If you dont know the answer, just say so.
+
+                here is the context:
+                {relevant_sections}
+            """,
+            },
+            {
+                "role" : "assistant",
+                "content" : "Namaste, Ask me any questions on Bhagavat Gita!"
+            }
+        ]
+    initial_state = State(messages=history + [{"role": "user", "content": message}])
+    print("initial_state = ", initial_state)
+    response = graph.invoke(initial_state)
+    return response["messages"][-1].content
+
+
+def main():
+    print("Hello from langgraph-demo!")
+    gr.ChatInterface(
+        chat,
+        type="messages",
+        title="Let's chat on Bhagavat Gita",
+        examples=[
+            "What does Gita say about Karma?",
+            "Why did God create this world?",
+            "What is the relationship between knowledge and action?",
+            "Who are friends and enemies per Gita?"
+        ],
+    ).launch()
+
+
+if __name__ == "__main__":
+    main()

pyproject.toml

@@ -0,0 +1,18 @@
+[project]
+name = "langgraph-demo"
+version = "0.1.0"
+description = "Add your description here"
+readme = "README.md"
+requires-python = ">=3.12"
+dependencies = [
+    "chromadb>=1.0.15",
+    "dotenv>=0.9.9",
+    "gradio>=5.38.0",
+    "ipython>=9.4.0",
+    "langchain>=0.3.26",
+    "langchain-community>=0.3.27",
+    "langchain-openai>=0.3.28",
+    "langgraph>=0.5.3",
+    "pydantic>=2.11.7",
+    "pypdf>=5.8.0",
+]

requirements.txt

@@ -0,0 +1,10 @@
+chromadb
+dotenv
+gradio
+ipython
+langchain
+langchain-community
+langchain-openai
+langgraph
+pydantic
+pypdf