File size: 502 Bytes
a87d450 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 | import h5py as h5
import json
import pandas as pd
import torch
from sentence_transformers import SentenceTransformer
torch.set_num_threads(60)
D = pd.read_json("red-pajamas-es.json.gz", lines=True)
#modelname = 'sentence-transformers/all-MiniLM-L6-v2'
modelname = "hiiamsid/sentence_similarity_spanish_es"
model = SentenceTransformer(modelname)
print(D.columns)
print("embeddings article")
embeddings = model.encode(D.text)
with h5.File("red-pajamas-es.h5", "a") as f:
f["text"] = embeddings
|