tuling-video-click-top3
图灵联邦视频点击预测大赛线上第三-【ctr, embedding, 穿越特征】
Stars
61
Ecosystems:
Jupyter Notebook
tuling-video-click-top3
2020-TURING-TOPIA-Video-Click-SINGLE-LightGBM-top3
=============================================================================================================== LightGBM
2020-
****https://www.turingtopia.com/competitionnew/detail/e4880352b6ef4f9f8f28e8f98498dbc4/sketch ****2019.11.11-2020.03.09 **** Freak luweihai **** ****:: https://pan.baidu.com/s/1YPtg4QyiAdhRAMoxjis_Gw : 0a3r
1.
train.csv
| id | ID | VARCHAR2(50) | 111376681 |
| target | VARCHAR2(50) | 10 | |
| timestamp | VARCHAR2(50) | NULL | |
| deviceid | ID | VARCHAR2(50) | id |
| newsid | ID | VARCHAR2(50) | id |
| guid | ID | VARCHAR2(50) | id |
| pos | VARCHAR2(50) | ||
| app_version | app | VARCHAR2(50) | app |
| device_vendor | VARCHAR2(50) | ||
| netmodel | VARCHAR2(50) | ||
| osversion | VARCHAR2(50) | ||
| lng | VARCHAR2(50) | ||
| lat | VARCHAR2(50) | ||
| device_version | VARCHAR2(50) | ||
| ts | ID |
test.csv
| id | ID | VARCHAR2(50) | test_1test_3653592 |
| deviceid | ID | VARCHAR2(50) | id |
| newsid | ID | VARCHAR2(50) | id |
| guid | ID | VARCHAR2(50) | id |
| pos | VARCHAR2(50) | ||
| app_version | app | VARCHAR2(50) | app |
| device_vendor | VARCHAR2(50) | ||
| netmodel | VARCHAR2(50) | ||
| osversion | VARCHAR2(50) | ||
| lng | VARCHAR2(50) | ||
| lat | VARCHAR2(50) | ||
| device_version | VARCHAR2(50) | ||
| ts | ID |
app.csv
| id | ID | VARCHAR2(50) | test_1test_3653592 |
| deviceid | ID | VARCHAR2(50) | id |
| applist deviceid | ID | VARCHAR2(50) | appappapp_1,app_2.. |
test.csv
| id | ID | VARCHAR2(50) | test_1test_3653592 |
| deviceid | ID | VARCHAR2(50) | id |
| guid | ID | VARCHAR2(50) | id |
| outertag | VARCHAR2(50) | ||
| tag | VARCHAR2(50) | outertag | |
| level | VARCHAR2(50) | ||
| personidentification | VARCHAR2(50) | 1 0 | |
| followscore | VARCHAR2(50) | ||
| personalscore | VARCHAR2(50) | ||
| gender | VARCHAR2(50) |
2.
- python3
- scikit-learn
- gensim
- Ubuntu
- LightGBM
- notebook
3.
1,2,3,4
1 feature.ipynb 2 fold_model.ipynb 3 offline_model.ipynb 4 online_model.ipynb
4.
-
-
-
- embedding
5.
0.83695