Libraries
","metadata":{}},{"cell_type":"code","source":"import pandas as pd\nimport numpy as np\n\nimport glob\n\n\nimport warnings\nwarnings.filterwarnings(\"ignore\")\n\nfrom datetime import datetime\n\nfrom tqdm import tqdm\n\ntype_labels = {'clicks':1, 'carts':2, 'orders':3}","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:07:05.146271Z","iopub.execute_input":"2024-06-08T08:07:05.147104Z","iopub.status.idle":"2024-06-08T08:07:06.373442Z","shell.execute_reply.started":"2024-06-08T08:07:05.147064Z","shell.execute_reply":"2024-06-08T08:07:06.372394Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"##Data Loading
","metadata":{}},{"cell_type":"code","source":"def load(which): \n dfs = []\n\n # 使用 glob 模块的 glob 函数来获取指定目录下所有符合模式的文件路径。这里的模式是 /kaggle/input/otto-chunk-data-inparquet-format/ 加上 which 参数和 _parquet/*，意味着它会查找所有以 _parquet 结尾的文件夹中的所有文件。\n test_files = glob.glob('/kaggle/input/otto-chunk-data-inparquet-format/'+which+'_parquet/*')\n \n for e, chunk_file in enumerate(test_files): # 枚举文件夹下所有文件\n chunk = pd.read_parquet(chunk_file)\n chunk.ts = (chunk.ts/1000).astype('int32') # chunk.ts ts为chunk下的一列，将其/1000后改为int32形式\n chunk['type'] = chunk['type'].map(type_labels).astype('int8') # type_labels = {'clicks':1, 'carts':2, 'orders':3} 将clicks,carts等字符串映射为整数1，2，3并改为int8类型\n \n dfs.append(chunk)\n\n \n return pd.concat(dfs).reset_index(drop=True) #.astype({\"ts\": \"datetime64[ms]\"})","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:07:06.375821Z","iopub.execute_input":"2024-06-08T08:07:06.376427Z","iopub.status.idle":"2024-06-08T08:07:06.385390Z","shell.execute_reply.started":"2024-06-08T08:07:06.376378Z","shell.execute_reply":"2024-06-08T08:07:06.383914Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#test_df = load('test')\ntrain_df = load('train')","metadata":{"_kg_hide-input":false,"_kg_hide-output":true,"execution":{"iopub.status.busy":"2024-06-08T08:07:06.387075Z","iopub.execute_input":"2024-06-08T08:07:06.387458Z","iopub.status.idle":"2024-06-08T08:08:05.373091Z","shell.execute_reply.started":"2024-06-08T08:07:06.387428Z","shell.execute_reply":"2024-06-08T08:08:05.370220Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_df","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:08:05.375909Z","iopub.execute_input":"2024-06-08T08:08:05.376353Z","iopub.status.idle":"2024-06-08T08:08:05.411827Z","shell.execute_reply.started":"2024-06-08T08:08:05.376317Z","shell.execute_reply":"2024-06-08T08:08:05.410613Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"##Generation of small dataset
","metadata":{}},{"cell_type":"markdown","source":"## To reduce runtime, only consider data from the last 4 days as train_df.","metadata":{}},{"cell_type":"code","source":"datetime.strptime(\"2022-08-25 00:00:00\", \"%Y-%m-%d %H:%M:%S\").timestamp()","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:08:05.417107Z","iopub.execute_input":"2024-06-08T08:08:05.418137Z","iopub.status.idle":"2024-06-08T08:08:05.427046Z","shell.execute_reply.started":"2024-06-08T08:08:05.418090Z","shell.execute_reply":"2024-06-08T08:08:05.425411Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sum(train_df['ts']>=1661385600) # 获得最近四天的数量","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:08:05.428664Z","iopub.execute_input":"2024-06-08T08:08:05.429059Z","iopub.status.idle":"2024-06-08T08:08:25.852709Z","shell.execute_reply.started":"2024-06-08T08:08:05.429023Z","shell.execute_reply":"2024-06-08T08:08:25.851601Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_df = train_df[train_df['ts']>=1661385600].reset_index(drop=True) # 选择最近4天的数量","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:08:25.854786Z","iopub.execute_input":"2024-06-08T08:08:25.855257Z","iopub.status.idle":"2024-06-08T08:08:27.713951Z","shell.execute_reply.started":"2024-06-08T08:08:25.855218Z","shell.execute_reply":"2024-06-08T08:08:27.712955Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_df","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:08:27.715255Z","iopub.execute_input":"2024-06-08T08:08:27.715584Z","iopub.status.idle":"2024-06-08T08:08:27.728939Z","shell.execute_reply.started":"2024-06-08T08:08:27.715557Z","shell.execute_reply":"2024-06-08T08:08:27.727535Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_df['datetime'] = train_df['ts'].apply(lambda x: datetime.fromtimestamp(x)) # 增加datatime列","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:08:27.730622Z","iopub.execute_input":"2024-06-08T08:08:27.731105Z","iopub.status.idle":"2024-06-08T08:09:11.997422Z","shell.execute_reply.started":"2024-06-08T08:08:27.731065Z","shell.execute_reply":"2024-06-08T08:09:11.996270Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_df ","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:11.999033Z","iopub.execute_input":"2024-06-08T08:09:11.999497Z","iopub.status.idle":"2024-06-08T08:09:12.022595Z","shell.execute_reply.started":"2024-06-08T08:09:11.999455Z","shell.execute_reply":"2024-06-08T08:09:12.020622Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_df.shape","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:12.024762Z","iopub.execute_input":"2024-06-08T08:09:12.025355Z","iopub.status.idle":"2024-06-08T08:09:12.038036Z","shell.execute_reply.started":"2024-06-08T08:09:12.025310Z","shell.execute_reply":"2024-06-08T08:09:12.036778Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_df['session'].nunique() # nunique() 函数用于计算DataFrame列中不同值的数量。当你调用 train_df['session'].nunique() 时，你正在计算 train_df DataFrame中 session 列中的唯一值有多少个。","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:12.039547Z","iopub.execute_input":"2024-06-08T08:09:12.039937Z","iopub.status.idle":"2024-06-08T08:09:12.893062Z","shell.execute_reply.started":"2024-06-08T08:09:12.039904Z","shell.execute_reply":"2024-06-08T08:09:12.891921Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_df['aid'].nunique()","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:12.894926Z","iopub.execute_input":"2024-06-08T08:09:12.895822Z","iopub.status.idle":"2024-06-08T08:09:13.611125Z","shell.execute_reply.started":"2024-06-08T08:09:12.895778Z","shell.execute_reply":"2024-06-08T08:09:13.609949Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_df","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:13.618045Z","iopub.execute_input":"2024-06-08T08:09:13.618413Z","iopub.status.idle":"2024-06-08T08:09:13.631376Z","shell.execute_reply.started":"2024-06-08T08:09:13.618382Z","shell.execute_reply":"2024-06-08T08:09:13.630177Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"##Division of training set and validation set
","metadata":{}},{"cell_type":"markdown","source":"## Choose the last day as the validation set.","metadata":{}},{"cell_type":"code","source":"# 总共就四天，因此自然是28，27，26，25这四天\ntrain_df['datetime'].dt.day.unique() # dt 是pandas中用于访问 datetime 列属性的属性访问器。.dt.day 则用于访问每个 datetime 对象的 day 属性，即日期中的天数部分。","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:13.632688Z","iopub.execute_input":"2024-06-08T08:09:13.633076Z","iopub.status.idle":"2024-06-08T08:09:14.620554Z","shell.execute_reply.started":"2024-06-08T08:09:13.633045Z","shell.execute_reply":"2024-06-08T08:09:14.619078Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_df['day'] = train_df['datetime'].dt.day # 增加day这一列","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:14.622078Z","iopub.execute_input":"2024-06-08T08:09:14.622532Z","iopub.status.idle":"2024-06-08T08:09:15.504661Z","shell.execute_reply.started":"2024-06-08T08:09:14.622493Z","shell.execute_reply":"2024-06-08T08:09:15.503616Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_df","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:15.505940Z","iopub.execute_input":"2024-06-08T08:09:15.506258Z","iopub.status.idle":"2024-06-08T08:09:15.521786Z","shell.execute_reply.started":"2024-06-08T08:09:15.506230Z","shell.execute_reply":"2024-06-08T08:09:15.520544Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_val_offline = train_df[train_df['day']==28].reset_index(drop=True) # 选择第28天作为验证集\ndf_train_offline = train_df[train_df['day']<28].reset_index(drop=True) # 其他的作为训练集","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:15.523667Z","iopub.execute_input":"2024-06-08T08:09:15.524520Z","iopub.status.idle":"2024-06-08T08:09:17.825807Z","shell.execute_reply.started":"2024-06-08T08:09:15.524473Z","shell.execute_reply":"2024-06-08T08:09:17.824652Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"##Recommendation based on item popularity
","metadata":{}},{"cell_type":"markdown","source":"## Analyze item popularity based on different types of behavior.","metadata":{}},{"cell_type":"code","source":"aid_pop_sta = df_train_offline.groupby(['aid','type'],as_index=False)['session'].count()","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:17.827061Z","iopub.execute_input":"2024-06-08T08:09:17.827397Z","iopub.status.idle":"2024-06-08T08:09:21.350495Z","shell.execute_reply.started":"2024-06-08T08:09:17.827371Z","shell.execute_reply":"2024-06-08T08:09:21.349320Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_train_offline","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:21.352074Z","iopub.execute_input":"2024-06-08T08:09:21.352498Z","iopub.status.idle":"2024-06-08T08:09:21.367938Z","shell.execute_reply.started":"2024-06-08T08:09:21.352464Z","shell.execute_reply":"2024-06-08T08:09:21.366715Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"aid_pop_sta\n","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:21.370056Z","iopub.execute_input":"2024-06-08T08:09:21.370517Z","iopub.status.idle":"2024-06-08T08:09:21.387061Z","shell.execute_reply.started":"2024-06-08T08:09:21.370477Z","shell.execute_reply":"2024-06-08T08:09:21.385885Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"aid_pop_sta.columns = ['aid','type','pop']","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:21.388781Z","iopub.execute_input":"2024-06-08T08:09:21.389167Z","iopub.status.idle":"2024-06-08T08:09:21.399802Z","shell.execute_reply.started":"2024-06-08T08:09:21.389136Z","shell.execute_reply":"2024-06-08T08:09:21.398700Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"aid_pop_sta","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:21.401126Z","iopub.execute_input":"2024-06-08T08:09:21.401465Z","iopub.status.idle":"2024-06-08T08:09:21.419538Z","shell.execute_reply.started":"2024-06-08T08:09:21.401437Z","shell.execute_reply":"2024-06-08T08:09:21.418154Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Obtain the top 200 most popular items.","metadata":{}},{"cell_type":"code","source":"aid_pop_sta = aid_pop_sta.sort_values(['type','pop'],ascending=False).reset_index(drop=True) # 排序和重置索引","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:12:19.407984Z","iopub.execute_input":"2024-06-08T08:12:19.408376Z","iopub.status.idle":"2024-06-08T08:12:19.610008Z","shell.execute_reply.started":"2024-06-08T08:12:19.408344Z","shell.execute_reply":"2024-06-08T08:12:19.608759Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"aid_pop_sta","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:12:22.084074Z","iopub.execute_input":"2024-06-08T08:12:22.084465Z","iopub.status.idle":"2024-06-08T08:12:22.097447Z","shell.execute_reply.started":"2024-06-08T08:12:22.084434Z","shell.execute_reply":"2024-06-08T08:12:22.096150Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"top200_aid_click = list(aid_pop_sta.loc[aid_pop_sta['type']==1,'aid'][:200])\ntop200_aid_cart = list(aid_pop_sta.loc[aid_pop_sta['type']==2,'aid'][:200])\ntop200_aid_order = list(aid_pop_sta.loc[aid_pop_sta['type']==3,'aid'][:200])\n\n'''\n这里使用的 .loc 方法是pandas中用于基于标签的索引和选择数据的方法。.loc[aid_pop_sta['type']==1,'aid'] 表示选择 aid_pop_sta DataFrame中 type 列值等于1的行的 aid 列。然后，[:200] 表示从筛选后的结果中取前200行。\n'''","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:12:30.379143Z","iopub.execute_input":"2024-06-08T08:12:30.379543Z","iopub.status.idle":"2024-06-08T08:12:30.415836Z","shell.execute_reply.started":"2024-06-08T08:12:30.379512Z","shell.execute_reply":"2024-06-08T08:12:30.414292Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Obtain the historical behavior list of the target user. The users in the validation set are the target users.","metadata":{}},{"cell_type":"code","source":"target_users = list( df_val_offline['session'].unique() )\ndf_val_offline['session'].unique() # 获得session中不同的数值，","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:12:32.474105Z","iopub.execute_input":"2024-06-08T08:12:32.474850Z","iopub.status.idle":"2024-06-08T08:12:32.853308Z","shell.execute_reply.started":"2024-06-08T08:12:32.474813Z","shell.execute_reply":"2024-06-08T08:12:32.852185Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"len(target_users)","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:12:34.653900Z","iopub.execute_input":"2024-06-08T08:12:34.654693Z","iopub.status.idle":"2024-06-08T08:12:34.662582Z","shell.execute_reply.started":"2024-06-08T08:12:34.654651Z","shell.execute_reply":"2024-06-08T08:12:34.661252Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"session_his = df_train_offline.loc[df_train_offline['session'].isin(target_users)].groupby(['session','type'],as_index=False)['aid'].agg(list)\n# 目的是收集特定用户（target_users）的会话历史中的aid值，并将这些值按会话和类型分组并聚合成列表。","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:12:43.224144Z","iopub.execute_input":"2024-06-08T08:12:43.224549Z","iopub.status.idle":"2024-06-08T08:13:02.727695Z","shell.execute_reply.started":"2024-06-08T08:12:43.224521Z","shell.execute_reply":"2024-06-08T08:13:02.726579Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"session_his # 观察列表我们可以发现，同一给session会有多个aid值，也就是对应有个aid列表","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:13:02.729596Z","iopub.execute_input":"2024-06-08T08:13:02.729950Z","iopub.status.idle":"2024-06-08T08:13:02.747434Z","shell.execute_reply.started":"2024-06-08T08:13:02.729920Z","shell.execute_reply":"2024-06-08T08:13:02.746217Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"session_his['session'].nunique()","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:13:02.748908Z","iopub.execute_input":"2024-06-08T08:13:02.749380Z","iopub.status.idle":"2024-06-08T08:13:02.800102Z","shell.execute_reply.started":"2024-06-08T08:13:02.749340Z","shell.execute_reply":"2024-06-08T08:13:02.798988Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**Only 558344 out of 1350502 target users have historical behavior.**","metadata":{}},{"cell_type":"code","source":"his_click_dict = dict(zip(session_his.loc[session_his['type']==1,'session'],session_his.loc[session_his['type']==1,'aid']))\nhis_cart_dict = dict(zip(session_his.loc[session_his['type']==2,'session'],session_his.loc[session_his['type']==2,'aid']))\nhis_order_dict = dict(zip(session_his.loc[session_his['type']==3,'session'],session_his.loc[session_his['type']==3,'aid']))","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:13:02.802487Z","iopub.execute_input":"2024-06-08T08:13:02.802835Z","iopub.status.idle":"2024-06-08T08:13:03.217332Z","shell.execute_reply.started":"2024-06-08T08:13:02.802804Z","shell.execute_reply":"2024-06-08T08:13:03.216190Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:13:03.218651Z","iopub.execute_input":"2024-06-08T08:13:03.219072Z","iopub.status.idle":"2024-06-08T08:13:03.484007Z","shell.execute_reply.started":"2024-06-08T08:13:03.219035Z","shell.execute_reply":"2024-06-08T08:13:03.482881Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Recommend to the target users based on item popularity, while excluding items that have appeared in the users' historical behavior records.","metadata":{}},{"cell_type":"code","source":"res = []\n\nfor each_session in tqdm(target_users):\n\n if each_session in his_click_dict:\n his_click_tmp = his_click_dict[each_session]\n res.append([str(each_session)+'_clicks'] + [' '.join([str(x) for x in top200_aid_click if x not in his_click_tmp][:20])] )\n else:\n res.append([str(each_session)+'_clicks'] + [' '.join([str(x) for x in top200_aid_click][:20])])\n \n \n if each_session in his_cart_dict:\n his_cart_tmp = his_cart_dict[each_session]\n res.append([str(each_session)+'_carts'] + [' '.join([str(x) for x in top200_aid_cart if x not in his_cart_tmp][:20])] )\n else:\n res.append([str(each_session)+'_carts'] + [' '.join([str(x) for x in top200_aid_cart][:20])])\n \n if each_session in his_order_dict:\n his_order_tmp = his_order_dict[each_session]\n res.append([str(each_session)+'_orders'] + [' '.join([str(x) for x in top200_aid_order if x not in his_order_tmp][:20])] )\n else:\n res.append([str(each_session)+'_orders'] + [' '.join([str(x) for x in top200_aid_order][:20])])\n ","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:13:03.485305Z","iopub.execute_input":"2024-06-08T08:13:03.485629Z","iopub.status.idle":"2024-06-08T08:16:08.756269Z","shell.execute_reply.started":"2024-06-08T08:13:03.485600Z","shell.execute_reply":"2024-06-08T08:16:08.754795Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_res = pd.DataFrame(res)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(df_res)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"res_dict = dict(zip(df_res[0],df_res[1]))","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"res_dict","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:21.534214Z","iopub.status.idle":"2024-06-08T08:09:21.534601Z","shell.execute_reply.started":"2024-06-08T08:09:21.534417Z","shell.execute_reply":"2024-06-08T08:09:21.534435Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"##Evaluate performance
","metadata":{}},{"cell_type":"markdown","source":"这是验证集，也就是最后的答案，官方给出的输出结果， 上面刚刚求的res_dict 是根据热度推给用户的产品","metadata":{}},{"cell_type":"code","source":"# 这是验证集，也就是最后的答案，官方给出的输出结果， 上面刚刚求的res_dict 是根据热度推给用户的产品\nsession_items_val = df_val_offline.groupby(['session','type'],as_index=False)['aid'].agg(set)","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:21.535830Z","iopub.status.idle":"2024-06-08T08:09:21.536246Z","shell.execute_reply.started":"2024-06-08T08:09:21.536060Z","shell.execute_reply":"2024-06-08T08:09:21.536076Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"session_items_val ","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:21.538215Z","iopub.status.idle":"2024-06-08T08:09:21.538595Z","shell.execute_reply.started":"2024-06-08T08:09:21.538417Z","shell.execute_reply":"2024-06-08T08:09:21.538434Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"session_items_val['session'].nunique()","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:21.540238Z","iopub.status.idle":"2024-06-08T08:09:21.540635Z","shell.execute_reply.started":"2024-06-08T08:09:21.540451Z","shell.execute_reply":"2024-06-08T08:09:21.540467Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"res_dict","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:21.541910Z","iopub.status.idle":"2024-06-08T08:09:21.542298Z","shell.execute_reply.started":"2024-06-08T08:09:21.542121Z","shell.execute_reply":"2024-06-08T08:09:21.542137Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def getScore(df, type_, suffix):\n df_tmp = df[df['type']==type_].reset_index(drop=True)\n print(df_tmp)\n score_Numerator = 0\n score_Denominator = 0\n\n for i,row in tqdm( df_tmp.iterrows(),total=len(df_tmp) ):\n recom_tmp = [int(x) for x in res_dict[str(row['session'])+suffix].split(' ')]\n \n score_Numerator += len(row['aid'] & set(recom_tmp))\n score_Denominator += min(20,len(row['aid']))\n \n return score_Numerator/score_Denominator","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:21.544545Z","iopub.status.idle":"2024-06-08T08:09:21.544936Z","shell.execute_reply.started":"2024-06-08T08:09:21.544738Z","shell.execute_reply":"2024-06-08T08:09:21.544752Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"score_click = getScore(session_items_val,1,'_clicks')","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:21.546181Z","iopub.status.idle":"2024-06-08T08:09:21.546573Z","shell.execute_reply.started":"2024-06-08T08:09:21.546378Z","shell.execute_reply":"2024-06-08T08:09:21.546393Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"score_cart = getScore(session_items_val,2,'_carts')","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:21.547791Z","iopub.status.idle":"2024-06-08T08:09:21.548216Z","shell.execute_reply.started":"2024-06-08T08:09:21.548023Z","shell.execute_reply":"2024-06-08T08:09:21.548039Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"score_order = getScore(session_items_val,3,'_orders')","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:21.550007Z","iopub.status.idle":"2024-06-08T08:09:21.550510Z","shell.execute_reply.started":"2024-06-08T08:09:21.550305Z","shell.execute_reply":"2024-06-08T08:09:21.550323Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"score_click * 0.1 + score_cart * 0.3 + score_order * 0.6","metadata":{"execution":{"iopub.status.busy":"2024-06-08T08:09:21.551629Z","iopub.status.idle":"2024-06-08T08:09:21.552040Z","shell.execute_reply.started":"2024-06-08T08:09:21.551829Z","shell.execute_reply":"2024-06-08T08:09:21.551845Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}