【Task03】组队学习新闻推荐系统—多路召回

【Task03】组队学习新闻推荐系统—多路召回

今天的代码比较难，时间比较紧，先码后面结合录播再看看细节，只跑通了一部分，部分注释在代码中。

import pandas as pd
import numpy as np
from tqdm import tqdm
from collections import defaultdict
import os, math, warnings, math, pickle
from tqdm import tqdm
import faiss
import collections
import random
from sklearn.preprocessing import MinMaxScaler
from sklearn.preprocessing import LabelEncoder
from datetime import datetime
from deepctr.feature_column import SparseFeat, VarLenSparseFeat
from sklearn.preprocessing import LabelEncoder
from tensorflow.python.keras import backend as K
from tensorflow.python.keras.models import Model
from tensorflow.python.keras.preprocessing.sequence import pad_sequences

from deepmatch.models import *
from deepmatch.utils import sampledsoftmaxloss
warnings.filterwarnings('ignore')

data_path = './data_raw/'
save_path = './temp_results/'
# 做召回评估的一个标志, 如果不进行评估就是直接使用全量数据进行召回
metric_recall = False


# debug模式： 从训练集中划出一部分数据来调试代码
def get_all_click_sample(data_path, sample_nums=10000):
    """
        训练集中采样一部分数据调试
        data_path: 原数据的存储路径
        sample_nums: 采样数目（这里由于机器的内存限制，可以采样用户做）
    """
    all_click = pd.read_csv(data_path + 'train_click_log.csv')
    all_user_ids = all_click.user_id.unique()

    sample_user_ids = np.random.choice(all_user_ids, size=sample_nums, replace=False)
    all_click = all_click[all_click['user_id'].isin(sample_user_ids)]

    all_click = all_click.drop_duplicates((['user_id', 'click_article_id', 'click_timestamp']))
    return all_click


# 读取点击数据，这里分成线上和线下，如果是为了获取线上提交结果应该讲测试集中的点击数据合并到总的数据中
# 如果是为了线下验证模型的有效性或者特征的有效性，可以只使用训练集
def get_all_click_df(data_path='./data_raw/', offline=True):
    if offline:
        all_click = pd.read_csv(data_path + 'train_click_log.csv')
    else:
        trn_click = pd.read_csv(data_path + 'train_click_log.csv')
        tst_click = pd.read_csv(data_path + 'testA_click_log.csv')

        all_click = trn_click.append(tst_click)

    all_click = all_click.drop_duplicates((['user_id', 'click_article_id', 'click_timestamp']))
    return all_click


# 读取文章的基本属性
def get_item_info_df(data_path):
    item_info_df = pd.read_csv(data_path + 'articles.csv')

    # 为了方便与训练集中的click_article_id拼接，需要把article_id修改成click_article_id
    item_info_df = item_info_df.rename(columns={
   'article_id': 'click_article_id'})

    return item_info_df


# 读取文章的Embedding数据
def get_item_emb_dict(data_path):
    item_emb_df = pd.read_csv(data_path + 'articles_emb.csv')

    item_emb_cols = [x for x in item_emb_df.columns if 'emb' in x]
    item_emb_np = np.ascontiguousarray(item_emb_df[item_emb_cols])
    # 进行归一化
    item_emb_np = item_emb_np / np.linalg.norm(item_emb_np, axis=1, keepdims=True)

    item_emb_dict = dict(zip(item_emb_df['article_id'], item_emb_np))
    pickle.dump(item_emb_dict, open(save_path + 'item_content_emb.pkl', 'wb'))

    return item_emb_dict

max_min_scaler = lambda x : (x-np.min(x))/(np.max(x)-np.min(x))

# 采样数据
# all_click_df = get_all_click_sample(data_path)

# 全量训练集
all_click_df = get_all_click_df(offline=False)

# 对时间戳进行归一化,用于在关联规则的时候计算权重
all_click_df['click_timestamp'] = all_click_df[['click_timestamp']].apply(max_min_scaler)

item_info_df = get_item_info_df(data_path)
item_emb_dict = get_item_emb_dict(data_path)


# 根据点击时间获取用户的点击文章序列   {user1: [(item1, time1), (item2, time2)..]...}
def get_user_item_time(click_df):
    click_df = click_df.sort_values('click_timestamp')

    def make_item_time_pair(df):
        return list(zip(df['click_article_id'], df['click_timestamp']))

    user_item_time_df = click_df.groupby('user_id')['click_article_id', 'click_timestamp'].apply(
        lambda x: make_item_time_pair(x)).reset_index().rename(columns={
   0: 'item_time_list'})
    user_item_time_dict = dict(zip(user_item_time_df['user_id'], user_item_time_df['item_time_list']))

    return user_item_time_dict

# 获取文章-用户-时间函数
# 根据时间获取商品被点击的用户序列  {item1: [(user1, time1), (user2, time2)...]...}
# 这里的时间是用户点击当前商品的时间，好像没有直接的关系。
def get_item_user_time_dict(click_df):
    def make_user_time_pair(df):
        return list(zip(df['user_id'], df['click_timestamp']))

    click_df = click_df.sort_values('click_timestamp')
    item_user_time_df = click_df.groupby('click_article_id')['user_id', 'click_timestamp'].apply(
        lambda x: make_user_time_pair(x)).reset_index().rename(columns={
   0: 'user_time_list'})

    item_user_time_dict = dict(zip(item_user_time_df['click_article_id'], item_user_time_df['user_time_list']))
    return item_user_time_dict


# 获取当前数据的历史点击和最后一次点击
def get_hist_and_last_click(all_click):
    all_click = all_click.sort_values(by=['user_id', 'click_timestamp'])
    click_last_df = all_click.groupby('user_id').tail(1) # 取用户最后一次点击

    # 如果用户只有一个点击，hist为空了，会导致训练的时候这个用户不可见，此时默认泄露一下
    def hist_func(user_df):
        if len(user_df) == 1:
            return user_df
        else:
            return user_df[:-1]

    click_hist_df = all_click.groupby('user_id').apply(hist_func).reset_index(drop=True) # 不保留原来索引

    return click_hist_df, click_last_df


# 获取文章id对应的基本属性，保存成字典的形式，方便后面召回阶段，冷启动阶段直接使用
def get_item_info_dict(item_info_df):
    max_min_scaler = lambda x: (x - np.min(x)) / (np.max(x) - np.min(x))
    item_info_df['created_at_ts'] = item_info_df[['created_at_ts']].apply(max_min_scaler) # 归一化

    item_type_dict = dict(zip(item_info_df['click_article_id'], item_info_df['category_id']))
    item_words_dict = dict(zip(item_info_df['click_article_id'], item_info_df['words_count']))
    item_created_time_dict = dict(zip(item_info_df['click_article_id'], item_info_df['created_at_ts']))

    return item_type_dict, item_words_dict, item_created_time_dict


def get_user_hist_item_info_dict(all_click):
    # 获取user_id对应的用户历史点击文章类型的集合字典
    user_hist_item_typs = all_click.groupby('user_id')['category_id'].agg(set).reset_index()
    user_hist_item_typs_dict = dict(zip(user_hist_item_typs['user_id'], user_hist_item_typs['category_id']))

    # 获取user_id对应的用户点击文章的集合
    user_hist_item_ids_dict = all_click.groupby('user_id')['click_article_id'].agg(set).reset_index()
    user_hist_item_ids_dict = dict(zip(user_hist_item_ids_dict['user_id'], user_hist_item_ids_dict['click_article_id']))

    # 获取user_id对应的用户历史点击的文章的平均字数字典
    user_hist_item_words = all_click.groupby('user_id')['words_count'].agg('mean').reset_index()
    user_hist_item_words_dict = dict(zip(user_hist_item_words['user_id'], user_hist_item_words['words_count']))

    # 获取user_id对应的用户最后一次点击的文章的创建时间
    all_click_ = all_click.sort_values('click_timestamp')
    user_last_item_created_time = all_click_.groupby('user_id')['created_at_ts'].apply(lambda x: x.iloc[-1]).reset_index()

    max_min_scaler = lambda x: (x - np.min(x)) / (np.max(x) - np.min(x))
    user_last_item_created_time['created_at_ts'] = user_last_item_created_time[['created_at_ts']].apply(max_min_scaler)# 点击时间归一化

    user_last_item_created_time_dict = dict(zip(user_last_item_created_time['user_id'],user_last_item_created_time['created_at_ts'