File size: 1,525 Bytes
05753e4 9e32d29 05753e4 4159dc2 8f10b0b 4159dc2 8f10b0b 4159dc2 8f10b0b 6cd48e0 4159dc2 d80f9fb 9e32d29 d80f9fb 4159dc2 9e32d29 17b5440 4159dc2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 |
import os
# If you want Gradio to run on a particular host/port, you can do this:
os.environ["GRADIO_SERVER_NAME"] = "0.0.0.0"
os.environ["GRADIO_SERVER_PORT"] = "7860"
os.environ["GRADIO_ROOT_PATH"] = "/_app/immutable"
import gradio as gr
from fastapi import FastAPI, Request
import uvicorn
import spaces
from sentence_transformers import SentenceTransformer
from sentence_transformers.util import cos_sim
from sentence_transformers.quantization import quantize_embeddings
app = FastAPI()
@spaces.GPU
def embed(text):
query_embedding = Embedder.encode(text)
return query_embedding.tolist();
@app.post("/v1/embeddings")
async def openai_embeddings(request: Request):
body = await request.json();
print(body);
model = body['model']
text = body['input'];
embeddings = embed(text)
return {
'object': "list"
,'data': [{
'object': "embeddings"
,'embedding': embeddings
,'index':0
}]
,'model':model
,'usage':{
'prompt_tokens': 0
,'total_tokens': 0
}
}
with gr.Blocks(fill_height=True) as demo:
text = gr.Textbox();
embeddings = gr.Textbox()
text.submit(embed, [text], [embeddings]);
print("Loading embedding model");
Embedder = None #SentenceTransformer("mixedbread-ai/mxbai-embed-large-v1")
GradioApp = gr.mount_gradio_app(app, demo, path="/", ssr_mode=False);
if __name__ == "__main__":
uvicorn.run(GradioApp, port=7860, host="0.0.0.0")
|