Add async lifespan context manager to preload model and update requirements for flexibility

main.py

@@ -11,6 +11,7 @@ from typing import List, Optional, Dict
 from datasets import Dataset, load_dataset
 from sentence_transformers import SentenceTransformer
 from huggingface_hub import login
+from contextlib import asynccontextmanager
 import pandas as pd
 import numpy as np
 import torch as t
@@ -22,6 +23,12 @@ from diskcache import Cache
 # Configure logging
 logging.basicConfig(level=logging.INFO)
 
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    # Preload the model
+    get_sentence_transformer()
+    yield
+    
 # Initialize FastAPI app
 app = FastAPI()
 

requirements.txt

@@ -1,14 +1,14 @@
-fastapi==0.109.2
-uvicorn==0.27.1
-python-jose==3.3.0
-python-multipart==0.0.6  # Required for OAuth2 form handling
-pydantic==2.6.1
-openai==1.12.0
-pandas==2.2.0
-numpy==1.26.3
-torch==2.1.2  # For sentence-transformers
-sentence-transformers==2.3.1
-datasets==2.17.0
-huggingface-hub==0.20.3
-diskcache==5.6.3
-python-dotenv==1.0.1  # For environment variable management
+fastapi
+uvicorn
+python-jose
+python-multipart  # Required for OAuth2 form handling
+pydantic
+openai
+pandas
+numpy
+torch  # For sentence-transformers
+sentence-transformers
+datasets
+huggingface-hub
+diskcache
+python-dotenv  # For environment variable management