app.py

"""
streamlit run app.py --server.address 0.0.0.0
"""

from __future__ import annotations

import os
from time import time

import faiss
import pandas as pd
import streamlit as st
from open_clip import create_model_and_transforms
from openai import OpenAI
from qdrant_client import QdrantClient
from qdrant_client.http import models

if os.getenv("SPACE_ID"):
    USE_HF_SPACE = True
    os.environ["HF_HOME"] = "/data/.huggingface"
    os.environ["HF_DATASETS_CACHE"] = "/data/.huggingface"
else:
    USE_HF_SPACE = False

# for tokenizer
os.environ["TOKENIZERS_PARALLELISM"] = "false"

OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY")
QDRANT_API_ENDPOINT = os.environ.get("QDRANT_API_ENDPOINT")
QDRANT_API_KEY = os.environ.get("QDRANT_API_KEY")

if not QDRANT_API_ENDPOINT or not QDRANT_API_KEY:
    raise ValueError("env: QDRANT_API_ENDPOINT or QDRANT_API_KEY is not set.")


@st.cache_resource
def get_model_preprocess():
    model, _, preprocess = create_model_and_transforms(
        "xlm-roberta-base-ViT-B-32", pretrained="laion5B-s13B-b90k"
    )
    return model, preprocess


@st.cache_resource
def get_qdrant_client():
    qdrant_client = QdrantClient(
        url=QDRANT_API_ENDPOINT,
        api_key=QDRANT_API_KEY,
    )
    return qdrant_client


def app():
    st.title("secon.dev site search")


if __name__ == "__main__":
    app()