7
7

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?

More than 5 years have passed since last update.

Pythonのベクトルの類似度計算高速化

Last updated at Posted at 2018-10-08

3万文書をベクトル化し、類似度計算するときに大量ベクトルの類似度計算にすさまじい時間がかかっていたので、効率的な計算方法を調査。scipyのライブラリを使うことで100倍くらい速くなったのでメモ。

calc_similarity.py
from scipy.spatial import distance
# ベクトルの作成 (Mは30,000×100の配列)
M = [a.vector for a in articles] # a.vectorは100要素のnp.arrayベクトル
# 30,000の記事の総当たり類似度を計算
dist_M = distance.cdist(M, M, metric='cosine')

結果がコサイン類似度ではなくコサイン距離で出るので注意。

今回の場合同じ配列同士を計算しているので、上三角形だけ計算するpdistの方がいいらしい。(計算時間が2分の1になる)。後ほど試す。

7
7
0

Register as a new user and use Qiita more conveniently

  1. You get articles that match your needs
  2. You can efficiently read back useful information
  3. You can use dark theme
What you can do with signing up
7
7

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?