Upload folder using huggingface_hub

.gitignore

@@ -0,0 +1,13 @@
+# Python
+__pycache__/
+*.py[cod]
+*.env
+
+# Logs
+*.log
+
+# Streamlit Configs
+*.stale
+
+# Other
+.DS_Store

README.md

@@ -1,12 +1,67 @@
----
-title: Diabetes Diet Bot
-emoji: 🦀
-colorFrom: purple
-colorTo: indigo
-sdk: gradio
-sdk_version: 5.12.0
-app_file: app.py
-pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+---
+title: Diabetes_Diet_Bot
+app_file: app.py
+sdk: gradio
+sdk_version: 5.12.0
+---
+# 🍛 **DiabetesDietBot** - South Indian Meal Planner for Type 2 Diabetes
+
+## **Introduction**
+DiabetesDietBot is an **AI-powered chatbot** designed to provide **personalized meal plans** for individuals with **Type 2 Diabetes**, focusing on **South Indian dietary preferences**. This chatbot leverages **Retrieval-Augmented Generation (RAG)** technology to retrieve relevant **meal plans** from a structured knowledge base (`MealPlans.pdf`), ensuring customized, nutritionally balanced recommendations.
+
+---
+
+## **🎯 Features**
+✔ **Personalized Meal Plans** – Based on **age, gender, dietary preference, and caloric needs**  
+✔ **South Indian Focus** – Meal plans aligned with traditional **vegetarian, non-vegetarian, and non-onion/garlic** diets  
+✔ **AI-Powered Recommendations** – Uses **FAISS indexing and OpenAI GPT** to provide the most relevant meal plans  
+✔ **Gradio-Based Chat Interface** – User-friendly, mobile-friendly chatbot for instant diet advice  
+✔ **Fast Retrieval** – FAISS-based vector search enables quick meal plan suggestions  
+
+---
+
+## **🛠 Technologies Used**
+### **Backend:**
+- **LangChain** – RAG-based retrieval and AI-assisted meal planning
+- **FAISS** – Vector search for quick and relevant meal plan retrieval
+- **OpenAI GPT API** – AI-powered chatbot responses
+
+### **Frontend:**
+- **Gradio** – Interactive chatbot UI with a sleek, modern interface
+
+### **Data Source:**
+- **MealPlans.pdf** – Contains 40+ structured meal plans tailored for Type 2 Diabetes in South Indian contexts
+
+---
+
+## **📂 Project Files**
+- `DiabetesDietBot.py` → Main chatbot script (Gradio-based UI)
+- `rag.py` → Text processing & vector indexing with FAISS
+- `rag2.py` → Alternate FAISS indexing method
+- `MealPlans.pdf` → Core dataset for AI meal plan recommendations
+- `requirements.txt` → Dependencies for Hugging Face deployment
+
+---
+
+## **🚀 Deployment on Hugging Face**
+You can use **Hugging Face Spaces** to deploy DiabetesDietBot:
+
+### **1️⃣ Manual Deployment**
+1. **Go to [Hugging Face Spaces](https://huggingface.co/spaces)**
+2. Click **"New Space"** → Choose **Gradio** as the SDK  
+3. **Upload the following files manually:**
+   - `DiabetesDietBot.py`
+   - `rag.py`
+   - `rag2.py`
+   - `MealPlans.pdf`
+   - `requirements.txt`
+4. **Restart Space** and your chatbot will be live!
+
+---
+
+## **🔧 Local Installation & Running**
+To run the chatbot locally:
+
+### **1️⃣ Install Dependencies**
+```bash
+pip install -r requirements.txt

app.py

@@ -0,0 +1,102 @@
+import gradio as gr
+import os
+from dotenv import load_dotenv
+from langchain_community.vectorstores import FAISS
+from langchain.embeddings import OpenAIEmbeddings
+import openai
+import PyMuPDF  # To process PDFs
+
+# Load environment variables
+load_dotenv()
+
+# Load and process the MealPlans.pdf
+pdf_path = "MealPlans.pdf"
+
+def load_pdf_text(pdf_path):
+    text = ""
+    with open(pdf_path, "rb") as f:
+        doc = PyMuPDF.open(f)
+        for page in doc:
+            text += page.get_text("text") + "\n"
+    return text
+
+if os.path.exists(pdf_path):
+    meal_plans_text = load_pdf_text(pdf_path)
+else:
+    raise FileNotFoundError("⚠️ Meal Plans PDF not found!")
+
+# Initialize OpenAI API Key
+openai_api_key = os.getenv("OPENAI_API_KEY")
+if not openai_api_key:
+    raise ValueError("⚠️ OpenAI API key not found! Please set the OPENAI_API_KEY environment variable.")
+
+openai.api_key = openai_api_key
+
+# Initialize FAISS Vector Store with the Meal Plans
+try:
+    embeddings = OpenAIEmbeddings()
+    vector_store = FAISS.from_texts([meal_plans_text], embedding=embeddings)
+except Exception as e:
+    raise Exception(f"⚠️ Error initializing FAISS vector store: {str(e)}")
+
+# Function to Retrieve Personalized Meal Plans
+def get_diet_plan(user_input):
+    """Retrieve personalized meal plans based on user input."""
+    try:
+        docs = vector_store.similarity_search(user_input, k=3)
+        meal_suggestions = '\n\n'.join([doc.page_content for doc in docs])
+
+        messages = [
+            {"role": "system", "content": "You are a diabetes dietitian providing personalized South Indian meal plans."},
+            {"role": "user", "content": f"Provide a meal plan based on this query: {user_input}\n\n{meal_suggestions}"}
+        ]
+
+        response = openai.ChatCompletion.create(
+            model="gpt-3.5-turbo",
+            messages=messages,
+            temperature=0.5
+        )
+
+        return response.choices[0].message['content']
+    except Exception as e:
+        return f"⚠️ Error generating diet plan: {str(e)}"
+
+# 🎨 Fancy Gradio UI with Custom Styling
+css = """
+h1 {
+    text-align: center;
+    color: #008000;
+    font-size: 2.5rem;
+}
+body {
+    background: linear-gradient(to right, #f3f4f6, #e0f7fa);
+}
+#chatbot {
+    font-size: 1.2rem;
+    background: #ffffff;
+    padding: 10px;
+    border-radius: 10px;
+    box-shadow: 0px 4px 10px rgba(0, 0, 0, 0.1);
+}
+.gradio-container {
+    max-width: 700px;
+    margin: auto;
+}
+"""
+
+# Gradio Interface
+with gr.Blocks(css=css) as app:
+    gr.Markdown("# 🍛 **DiabetesDietBot** - South Indian Meal Planner for Type 2 Diabetes")
+
+    gr.Markdown("👋 Welcome! Enter details like **'Male, 40s, Vegetarian, Moderate Calories'** to get a **customized diabetes-friendly meal plan!**")
+
+    chatbot = gr.ChatInterface(fn=get_diet_plan, chatbot=True, height=400, placeholder="Enter your dietary needs...")
+
+    with gr.Row():
+        img = gr.Image("meal_plan_example.jpg", width=400, height=250, interactive=False)
+        gr.Markdown("### Example: South Indian Diabetes-Friendly Meal Plan")
+
+    gr.Markdown("### 🔗 Developed by [Dr. Syed Faizan](https://www.linkedin.com/in/drsyedfaizanmd/)")
+
+# Launch Gradio App
+app.launch()

db_mealplans/index.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd6841681ca462d9b4c8b7e9bd7be999efdcda2a1bfc3f4d00314abac116b079
+size 4483

desktop.ini

@@ -0,0 +1,10 @@
+[LocalizedFileNames]
+consolidatedengagedcitizenry.xlsx=@consolidatedengagedcitizenry,0
+consolidatedhousing.xlsx=@consolidatedhousing,0
+consolidatedcommunityservicesdata.xlsx=@consolidatedcommunityservicesdata,0
+Housing Stability Dashboard 2.pdf=@Housing Stability Dashboard 2,0
+Housing Stability Dashboard 1.pdf=@Housing Stability Dashboard 1,0
+Financial_Stability_Dashboard.pdf=@Financial_Stability_Dashboard,0
+Engaged and Connected Residents Dashboard.pdf=@Engaged and Connected Residents Dashboard,0
+Demographics Dashboard.pdf=@Demographics Dashboard,0
+Community Social Services Dashboard.pdf=@Community Social Services Dashboard,0

rag.py

@@ -0,0 +1,86 @@
+# %% Packages
+import os
+import re
+from dotenv import load_dotenv
+from pypdf import PdfReader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_community.vectorstores import FAISS
+from langchain_openai import OpenAIEmbeddings
+import openai
+
+# Load environment variables from .env file
+load_dotenv()
+
+# Ensure OpenAI API key is set
+openai_api_key = os.getenv("OPENAI_API_KEY")
+if not openai_api_key:
+    raise ValueError("⚠️ OpenAI API Key is missing. Set it in your .env file or environment variables.")
+
+# Create OpenAI client (✅ Fix for v1.x API)
+client = openai.OpenAI(api_key=openai_api_key)
+
+# Load the Meal Plans PDF
+pdf_path = "MealPlans.pdf"
+
+if not os.path.exists(pdf_path):
+    raise FileNotFoundError("⚠️ Meal Plans PDF not found!")
+
+# Extract text from PDF
+reader = PdfReader(pdf_path)
+meal_texts = [page.extract_text().strip() for page in reader.pages if page.extract_text()]
+
+# Clean extracted text
+cleaned_texts = [re.sub(r'\d+\n.*?\n', '', text) for text in meal_texts]
+
+# Split text into chunks
+char_splitter = RecursiveCharacterTextSplitter(
+    separators=["\n\n", "\n", ". ", " ", ""],
+    chunk_size=500,
+    chunk_overlap=50
+)
+
+# Ensure `document_chunks` is a **flat list**
+document_chunks = char_splitter.split_text("\n\n".join(cleaned_texts))  # Convert to a single string first
+print(f"✅ Number of text chunks: {len(document_chunks)}")
+
+# Generate embeddings
+embeddings = OpenAIEmbeddings()
+
+# Ensure FAISS receives a list of strings (not nested lists)
+vector_store = FAISS.from_texts(document_chunks, embedding=embeddings)
+
+# Save FAISS index
+vector_store.save_local("db_mealplans")
+print("✅ FAISS index successfully created and saved for DiabetesDietBot.")
+
+# Define RAG Query Function
+def rag(query, n_results=5):
+    """Retrieve personalized meal plans based on dietary preferences."""
+    try:
+        # Query FAISS vector store
+        docs = vector_store.similarity_search(query, k=n_results)
+        retrieved_text = "; ".join([doc.page_content for doc in docs])
+
+        # Prepare AI prompt
+        messages = [
+            {"role": "system", "content": "You are a diabetes dietitian specializing in South Indian meal planning."},
+            {"role": "user", "content": f"Generate a personalized meal plan: {query}\n\n{retrieved_text}"}
+        ]
+
+        # ✅ Fixed OpenAI API Call (New v1.x API)
+        response = client.chat.completions.create(
+            model="gpt-3.5-turbo",
+            messages=messages
+        )
+        
+        return response.choices[0].message.content
+
+    except Exception as e:
+        return f"⚠️ Error generating diet plan: {str(e)}"
+
+# Example Query Test
+if __name__ == "__main__":
+    test_query = "Male, 50s, Vegetarian, Moderate Calories"
+    response = rag(query=test_query, n_results=5)
+    print("\n🛑 Sample Response:\n")
+    print(response)

rag2.py

@@ -0,0 +1,34 @@
+from langchain_community.vectorstores import FAISS
+from langchain_openai.embeddings import OpenAIEmbeddings
+from PyPDF2 import PdfReader
+
+
+from dotenv import load_dotenv
+load_dotenv()
+
+# Function to extract text from PDF
+def extract_text_from_pdf(pdf_path):
+    reader = PdfReader(pdf_path)
+    text = ""
+    for page in reader.pages:
+        text += page.extract_text()
+    return text
+
+# Path to the uploaded document
+pdf_path = r"C:\Users\sfaiz\OneDrive\Desktop\Official ALY 6080 Website\ALY_6080_Experential_learning_Group_1_Module_12_Capstone_Sponsor_Deliverable.pdf"
+
+# Extract text from the PDF
+document_text = extract_text_from_pdf(pdf_path)
+
+# Split the document into smaller chunks for FAISS
+chunk_size = 1000  # Adjust based on requirements
+document_chunks = [document_text[i:i+chunk_size] for i in range(0, len(document_text), chunk_size)]
+
+# Generate embeddings
+embeddings = OpenAIEmbeddings()
+
+# Create FAISS index
+vector_store = FAISS.from_texts(document_chunks, embeddings)
+
+# Save the index locally in the 'db' folder
+vector_store.save_local("db")

requirements.txt

@@ -0,0 +1,39 @@
+# Hugging Face & Gradio
+huggingface_hub==0.25.2
+gradio==5.7.1
+
+# FAISS for Vector Search
+faiss-cpu==1.9.0
+
+# LangChain for Retrieval-Augmented Generation (RAG)
+langchain==0.3.15 
+langchain-community==0.3.7
+langchain-huggingface==0.1.2
+langchain-openai==0.2.10
+
+# OpenAI API for LLM Responses
+openai==1.59.9
+
+
+# Data Handling & Processing
+numpy==1.26.4
+pydantic==2.9.2
+python-dotenv==1.0.1
+
+# PDF Processing
+PyMuPDF==1.21.1  # Make sure this version is supported
+
+# Search Functionality (Optional for Expanding Queries)
+duckduckgo-search==2.9.0
+
+# Interactive Shell for Debugging (Optional)
+ipython==8.29.0
+
+# File Type Detection (Optional)
+python-magic==0.4.27
+
+
+
+
+
+