tag_data/cal.py

import numpy as np
from sentence_transformers import SentenceTransformer


encoder = SentenceTransformer("hfl/chinese-roberta-wwm-ext-large")

tags = []
f = open('tags.txt', 'r')
for line in f.readlines():
    tags.append(line.strip())
f.close()

tags_embed = encoder.encode(tags)
tags_dis = [np.sqrt(np.dot(_, _.T)) for _ in tags_embed]

print(tags_embed.shape, tags_dis.shape)

with open('./tags_embed.npy', 'wb') as f:
    np.save(f, tags_embed)

with open('./tags_dis.npy', 'wb') as f:
    np.save(f, tags_dis)