Delete upload_pdf.py
Browse files- upload_pdf.py +0 -26
upload_pdf.py
DELETED
@@ -1,26 +0,0 @@
|
|
1 |
-
import os
|
2 |
-
from fastapi import FastAPI, File, UploadFile
|
3 |
-
from langchain_community.document_loaders import PyMuPDFLoader
|
4 |
-
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
5 |
-
from langchain_community.embeddings import HuggingFaceEmbeddings
|
6 |
-
from langchain_community.vectorstores import Chroma
|
7 |
-
|
8 |
-
app = FastAPI()
|
9 |
-
|
10 |
-
@app.post("/upload")
|
11 |
-
async def upload_pdf(file: UploadFile = File(...)):
|
12 |
-
file_location = f"uploads/{file.filename}"
|
13 |
-
with open(file_location, "wb") as f:
|
14 |
-
f.write(await file.read())
|
15 |
-
|
16 |
-
loader = PyMuPDFLoader(file_location)
|
17 |
-
docs = loader.load()
|
18 |
-
|
19 |
-
splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
|
20 |
-
chunks = splitter.split_documents(docs)
|
21 |
-
|
22 |
-
embedding = HuggingFaceEmbeddings(model_name="all-MiniLM-L6-v2")
|
23 |
-
vectordb = Chroma.from_documents(chunks, embedding, persist_directory="chroma_db")
|
24 |
-
vectordb.persist()
|
25 |
-
|
26 |
-
return {"message": f"✅ {file.filename} uploaded and processed into vector DB."}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|