File size: 1,438 Bytes
d1659f7 1dd5fed d1659f7 12bb371 d1659f7 1dd5fed ac5a31f 1dd5fed |
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 |
from fastapi import FastAPI
from fastapi.staticfiles import StaticFiles
from fastapi.responses import FileResponse
from transformers import AutoTokenizer, AutoModel
import numpy as np
from sklearn.metrics.pairwise import cosine_similarity
# load model and tokenizer
tokenizer = AutoTokenizer.from_pretrained('allenai/specter')
model = AutoModel.from_pretrained('allenai/specter')
# papers = [{'title': 'BERT', 'abstract': 'We introduce a new language representation model called BERT'},
# {'title': 'Attention is all you need', 'abstract': ' The dominant sequence transduction models are based on complex recurrent or convolutional neural networks'}]
# concatenate title and abstract
app = FastAPI()
app.mount("/", StaticFiles(directory="static", html=True), name="static")
@app.get("/")
def index() -> FileResponse:
return FileResponse(path="/app/static/index.html", media_type="text/html")
@app.get('/similarity')
def similarity(input):
papers = input['papers']
title_abs = [d['title'] + tokenizer.sep_token + (d.get('abstract') or '') for d in papers]
# preprocess the input
inputs = tokenizer(title_abs, padding=True, truncation=True, return_tensors="pt", max_length=512)
result = model(**inputs)
# take the first token in the batch as the embedding
embeddings = result.last_hidden_state[:, 0, :]
res = cosine_similarity(embeddings, embeddings).tolist()
return {"output": res} |