Upload 5 files

.gitattributes

@@ -33,3 +33,5 @@ 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
+data/Attention.pdf filter=lfs diff=lfs merge=lfs -text
+data/VT.pdf filter=lfs diff=lfs merge=lfs -text

api.py

@@ -0,0 +1,76 @@
+from fastapi import FastAPI
+from langchain.document_loaders import DirectoryLoader, PyPDFLoader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain.embeddings import HuggingFaceEmbeddings
+from langchain.vectorstores import FAISS
+from langchain_huggingface import HuggingFaceEndpoint
+from langchain.chains.combine_documents import create_stuff_documents_chain
+from langchain_core.prompts import ChatPromptTemplate
+from langchain.chains import create_retrieval_chain
+import os
+import uvicorn
+
+# Initialize FastAPI App 🚀
+app = FastAPI(title="Vision Transformer Assistant", description="A FastAPI-powered AI assistant for deep learning.")
+
+# Load Hugging Face Token 🔑
+HF_TOKEN = os.getenv("HF_TOKEN")  # Ensure your HF_TOKEN is set in Hugging Face Secrets
+
+# Load Documents 📂
+loader = DirectoryLoader("./data/", glob="*.pdf", loader_cls=PyPDFLoader)
+docs = loader.load()
+
+# Text Splitting 📖
+text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
+texts = text_splitter.split_documents(docs)
+
+# Vector Database 🔍
+db = FAISS.from_documents(documents=texts, embedding=HuggingFaceEmbeddings(model_name='BAAI/bge-base-en-v1.5'))
+retriever = db.as_retriever()
+
+# Load LLM 🚀
+repo_id = "mistralai/Mistral-7B-Instruct-v0.3"
+llm = HuggingFaceEndpoint(repo_id=repo_id, token=HF_TOKEN, task="text-generation")
+
+# Prompt Template ✨
+prompt_temp = ChatPromptTemplate.from_template("""
+You are an AI assistant specializing in deep learning, specifically Vision Transformers.
+
+<context>
+{context}
+<context>
+
+### Instructions:
+- Extract relevant information only from retrieved documents.
+- Provide concise yet detailed responses.
+- Use LaTeX for equations when necessary.
+- Do not make up answers; respond with *'Information not available in retrieved documents.'* if needed.
+""")
+
+# Create Retrieval Chain ⚡
+document_chain = create_stuff_documents_chain(llm, prompt_temp)
+retrieval_chain = create_retrieval_chain(retriever, document_chain)
+
+def get_response(query: str) -> str:
+    """
+    Get AI-generated response based on query.
+    """
+    response = retrieval_chain.invoke({'input': query})
+    return response['answer']
+
+@app.get("/")
+def home():
+    return {"message": "Vision Transformer Assistant API is running 🚀"}
+
+@app.get("/query")
+def get_answer(query: str):
+    """
+    API endpoint to retrieve AI-generated responses.
+    """
+    answer = get_response(query)
+    return {"answer": answer}
+
+# Run FastAPI on Hugging Face Spaces (public)
+if __name__ == "__main__":
+    uvicorn.run(app, host="0.0.0.0", port=8080)
+

app.py

@@ -0,0 +1,28 @@
+import streamlit as st
+import requests
+
+# Hugging Face Spaces FastAPI Backend URL
+FASTAPI_URL = "https://pranavreddy18-research.hf.space/query"  # Change to your actual Space URL
+
+# Streamlit UI
+st.set_page_config(page_title="Vision Transformer Assistant", page_icon="🤖")
+st.title("Vision Transformer Assistant 🤖")
+st.markdown("Ask anything about deep learning and Vision Transformers!")
+
+# User input
+query = st.text_input("Enter your question:")
+
+if st.button("Get Answer"):
+    if query:
+        with st.spinner("Fetching answer..."):
+            try:
+                response = requests.get(FASTAPI_URL, params={"query": query})
+                if response.status_code == 200:
+                    answer = response.json().get("answer", "No answer found.")
+                    st.success("✅ Answer:")
+                    st.write(answer)
+                else:
+                    st.error(f"⚠️ Error fetching answer. Status Code: {response.status_code}")
+            except requests.exceptions.RequestException as e:
+                st.error(f"⚠️ Failed to connect to backend: {e}")
+

data/Attention.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bdfaa68d8984f0dc02beaca527b76f207d99b666d31d1da728ee0728182df697
+size 2215244

data/VT.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ce7b83971a14508ca711a27c875c9b6914c4f6767cf3150fb1ca6c07aa056d6
+size 3743814