def __get_lfm_matrix(self, lfm_matrix_path):
try:
print("开始载入隐语义矩阵....")
self.user_p, self.movie_q = load_file(lfm_matrix_path)
print("载入隐语义矩阵完成")
except FileNotFoundError:
print("载入隐语义矩阵失败，重新计算隐语义矩阵")
# 建立隐语义矩阵
self.user_p, self.movie_q = self.__build_matrix()
print("开始保存隐语义矩阵")
save_file(lfm_matrix_path, (self.user_p, self.movie_q))
print("保存隐语义矩阵完成")
def train(self, origin_data, sim_matrix_path="store/user_sim.pkl"):
"""训练模型
@param origin_data: 原始数据
@param sim_matrix_path: 协同矩阵保存的路径
"""
# 初始化训练集
self._init_train(origin_data)
print("开始训练模型", file=sys.stderr)
try:
print("开始载入用户协同矩阵....", file=sys.stderr)
self.user_sim_matrix = load_file(sim_matrix_path)
print("载入协同过滤矩阵完成", file=sys.stderr)
except BaseException:
print("载入用户协同过滤矩阵失败，重新计算协同过滤矩阵", file=sys.stderr)
# 计算用户协同矩阵
self.user_sim_matrix = self.user_similarity()
print("开始保存协同过滤矩阵", file=sys.stderr)
save_file(sim_matrix_path, self.user_sim_matrix)
print("保存协同过滤矩阵完成", file=sys.stderr)
def train(self, sim_matrix_path="store/item_sim.pkl"):
"""
训练模型
:param sim_matrix_path: 协同矩阵保存的路径
:return:
"""
# 初始化训练集
UserCF._init_train(self)
print("开始训练模型", file=sys.stderr)
try:
print("开始载入用户协同矩阵....", file=sys.stderr)
self.item_sim_matrix = load_file(sim_matrix_path)
print("载入协同过滤矩阵完成", file=sys.stderr)
except FileNotFoundError:
print("载入用户协同过滤矩阵失败，重新计算协同过滤矩阵", file=sys.stderr)
# 计算用户协同矩阵
self.item_sim_matrix = self._item_similarity()
print("开始保存协同过滤矩阵", file=sys.stderr)
save_file(sim_matrix_path, self.item_sim_matrix)
print("保存协同过滤矩阵完成", file=sys.stderr)
from flask import Flask, jsonify
from main.util.utils import load_file
from main.LFM import LFM
import numpy as np
app = Flask(__name__)
lfm = LFM(10)
lfm.read_data()
item_pool, user_item=lfm.init_data(lfm.trainset)
user_p = load_file("store/lfm_user.pkl")
movie_q = load_file("store/lfm_movie.pkl")
@app.route('/')
def hello_world():
return 'Hello World!'
@app.route('/recom/<int:userId>')
def show_post(userId):
movie_rating=dict()
for movieId in item_pool.keys():
if movieId in user_item[userId]:
continue
movie_rating[movieId] = np.dot(user_p[userId], movie_q[movieId])
movie_rating = dict(sorted(movie_rating.items(), key=lambda x: x[1], reverse=True)[:10])
ret = list(movie_rating.keys())