Merge pull request #4 from Group-Project-II/dev

feat:add prompt api

app/__init__.py

@@ -2,9 +2,11 @@
 from .main import app
 
 from .routers.upload import router as upload_router
+from .routers.questionGenerate import router as questionGenerate_router
 
 
 app.include_router(upload_router, prefix="/api/v1")
+app.include_router(questionGenerate_router, prefix="/api/v1")
 
 
 # app/routers/upload.py

app/routers/questionGenerate.py

@@ -0,0 +1,14 @@
+from fastapi import APIRouter, Query, HTTPException
+from typing import List
+
+from ..services.prompt import prompt
+
+
+router = APIRouter()
+
+@router.get("/generate-question/", response_model=str)
+async def generate_question(text: str = Query(..., description="The text to generate a question for"),
+                            examid: str = Query(..., description="The ID of the exam related to the text")) -> str:
+    """Endpoint to generate a question for a given text using OpenAI's model."""
+
+    return prompt(text, examid) 

app/routers/upload.py

@@ -1,18 +1,20 @@
-from fastapi import APIRouter, File, UploadFile, HTTPException, Response
+from fastapi import APIRouter, File, UploadFile, HTTPException, Response, Query
 import os
 
 from ..services.pinecone_upsert import upsert
 
 router = APIRouter()
 
 @router.post("/upload-pdf/", status_code=201)
-async def upload_pdf(file: UploadFile = File(...)) -> dict:
+async def upload_pdf(file: UploadFile = File(...), examid:str = Query(..., description="The ID of the exam related to the uploaded PDF") ) -> dict:
     """Endpoint to upload a PDF and upsert its contents into a Pinecone vector store."""
 
     if file.content_type != 'application/pdf':
         raise HTTPException(status_code=415, detail="Unsupported file type. Please upload a PDF.")
 
     # Call the upsert function from the imported service
-    upsert(file, examid='examid')
+    upsert(file, examid)
 
-    return {"filename": file.filename}
+    # return {"filename": file.filename}
+    Response(status_code=201)
+    return {"message": "PDF uploaded successfully."}

app/services/prompt.py

@@ -0,0 +1,64 @@
+from typing import Any, BinaryIO
+import os
+import dotenv
+import pdfplumber
+from langchain_openai import OpenAIEmbeddings
+from langchain_pinecone import PineconeVectorStore
+from pinecone import Pinecone, ServerlessSpec
+
+from langchain.chat_models import ChatOpenAI
+from langchain.chains import RetrievalQA
+
+
+
+dotenv.load_dotenv()
+
+pinecone = Pinecone(api_key=os.getenv('PINECONE_API_KEY'))
+
+
+
+
+def prompt(text: str, examid: str) -> str:
+    """Upserts PDF text into a Pinecone vector store and returns the extracted text."""
+
+    embed = OpenAIEmbeddings(
+        model="text-embedding-3-large",
+        api_key=os.getenv('OPENAI_API_KEY'),
+        dimensions=3072
+    )
+
+    vectorstore = PineconeVectorStore(
+        namespace=examid,
+        index_name="abc",
+        embedding=embed
+
+    )
+
+    vectorstore.similarity_search(
+        text,
+        # top_k=5
+    )
+
+    llm = ChatOpenAI(
+        model="gpt-3.5-turbo",
+        api_key=os.getenv('OPENAI_API_KEY')
+    )
+
+    qa = RetrievalQA.from_chain_type(
+        llm=llm,
+        chain_type="stuff",
+        retriever=vectorstore.as_retriever()
+    )
+
+    print(qa.invoke(text))
+    return "Question generated successfully."
+
+
+
+
+
+
+
+
+
+