基于语义的泛化系统——annoy近邻搜索实现

数据预处理和二叉树构建

build.py

# 基于annoy二叉树的近邻搜索版本 构建

import json
import numpy as np
from collections import OrderedDict
from annoy import AnnoyIndex


def build(word_path='../../res/res/word1.txt', vec_path='../../res/res/vector1.npy', build_trees=1, annoy_deep=200):
    keywords = getKeyword(word_path)
    vectors = getVectors(vec_path)
    word_index = OrderedDict()
    # 关键词->索引 表
    for count, word in enumerate(keywords)