All Projects → shenweichen → Tencent_social_ads2017_mobile_app_pcvr

shenweichen / Tencent_social_ads2017_mobile_app_pcvr

Tencent Social Ads 2017 contest rank 20

Projects that are alternatives of or similar to Tencent social ads2017 mobile app pcvr

Tensorflow Multi Dimensional Lstm
Multi dimensional LSTM as described in Alex Graves' Paper https://arxiv.org/pdf/0705.2011.pdf
Stars: ✭ 154 (-0.65%)
Mutual labels:  jupyter-notebook
Pyportfolioopt
Financial portfolio optimisation in python, including classical efficient frontier, Black-Litterman, Hierarchical Risk Parity
Stars: ✭ 2,502 (+1514.19%)
Mutual labels:  jupyter-notebook
Spiking Neural Network Snn With Pytorch Where Backpropagation Engenders Stdp
What about coding a Spiking Neural Network using an automatic differentiation framework? In SNNs, there is a time axis and the neural network sees data throughout time, and activation functions are instead spikes that are raised past a certain pre-activation threshold. Pre-activation values constantly fades if neurons aren't excited enough.
Stars: ✭ 155 (+0%)
Mutual labels:  jupyter-notebook
Matplotlib Label Lines
Label line using matplotlib.
Stars: ✭ 154 (-0.65%)
Mutual labels:  jupyter-notebook
Neural Style Transfer
Keras Implementation of Neural Style Transfer from the paper "A Neural Algorithm of Artistic Style" (http://arxiv.org/abs/1508.06576) in Keras 2.0+
Stars: ✭ 2,000 (+1190.32%)
Mutual labels:  jupyter-notebook
Davsod
Shifting More Attention to Video Salient Objection Detection, CVPR 2019 (Best paper finalist & Oral)
Stars: ✭ 155 (+0%)
Mutual labels:  jupyter-notebook
Example Seldon
Example for end-to-end machine learning on Kubernetes using Kubeflow and Seldon Core
Stars: ✭ 154 (-0.65%)
Mutual labels:  jupyter-notebook
Copulas
A library to model multivariate data using copulas.
Stars: ✭ 149 (-3.87%)
Mutual labels:  jupyter-notebook
Deepreinforcementlearning
A replica of the AlphaZero methodology for deep reinforcement learning in Python
Stars: ✭ 1,898 (+1124.52%)
Mutual labels:  jupyter-notebook
Pytorchmedicalai
This is the hands-on deep learning tutorial series for the 2018/2019 Medical AI course by DeepOncology AI.
Stars: ✭ 155 (+0%)
Mutual labels:  jupyter-notebook
Stock Market Prediction Challenge
Following repo is the solution to Stock Market Prediction using Neural Networks and Sentiment Analysis
Stars: ✭ 154 (-0.65%)
Mutual labels:  jupyter-notebook
Binderhub
Run your code in the cloud, with technology so advanced, it feels like magic!
Stars: ✭ 2,050 (+1222.58%)
Mutual labels:  jupyter-notebook
Your First Kaggle Submission
How to perform an exploratory data analysis on the Kaggle Titanic dataset and make a submission to the leaderboard.
Stars: ✭ 155 (+0%)
Mutual labels:  jupyter-notebook
Cnnvis Pytorch
visualization of CNN in PyTorch
Stars: ✭ 154 (-0.65%)
Mutual labels:  jupyter-notebook
Surgery Robot Detection Segmentation
Object detection and segmentation for a surgery robot using Mask-RCNN on Python 3, Keras, and TensorFlow..
Stars: ✭ 155 (+0%)
Mutual labels:  jupyter-notebook
Deep Viz Keras
Implementations of some popular Saliency Maps in Keras
Stars: ✭ 154 (-0.65%)
Mutual labels:  jupyter-notebook
Jupyter Vim Binding
Jupyter meets Vim. Vimmer will fall in love.
Stars: ✭ 1,965 (+1167.74%)
Mutual labels:  jupyter-notebook
Fcn For Semantic Segmentation
Implemention of FCN-8 and FCN-16 with Keras and uses CRF as post processing
Stars: ✭ 155 (+0%)
Mutual labels:  jupyter-notebook
Mgwr
Multiscale Geographically Weighted Regression (MGWR)
Stars: ✭ 155 (+0%)
Mutual labels:  jupyter-notebook
Stocks
Programs for stock prediction and evaluation
Stars: ✭ 155 (+0%)
Mutual labels:  jupyter-notebook

第一届腾讯社交广告高校算法大赛-移动App广告转化率预估

赛题详情http://algo.tpai.qq.com/home/information/index.html  
题目描述
根据从某社交广告系统连续两周的日志记录按照推广中的App和用户维度随机采样构造的数据,预测App广告点击后被激活的概率:pCVR=P(conversion=1 | Ad,User,Context),即给定广告、用户和上下文情况下广告被点击后发生激活的概率。

运行环境

  • 操作系统 Ubuntu 14.04.4 LTS (GNU/Linux 4.2.0-27-generic x86_64)
  • 内存 128GB
  • CPU 32 Intel(R) Xeon(R) CPU E5-2620 v4 @ 2.10GHz
  • 显卡 TITAN X (Pascal) 12GB
  • 语言 Python3.6
  • Python依赖包
    1. Keras==2.0.6
    2. lightgbm==0.1
    3. matplotlib==2.0.0
    4. numpy==1.11.3
    5. pandas==0.19.2
    6. scikit-learn==0.18.1
    7. scipy==0.18.1
    8. tensorflow-gpu==1.2.1
    9. tqdm==4.11.2
    10. xgboost==0.6a2
  • 其他库 LIBFFM v121

运行说明

  1. 将复赛数据文件final.zip放在根目录下
  2. 在根目录下运行sh run.sh命令生成特征文件
  3. 打开./code/_4_*_model_*.ipynb分别进行模型训练和预测,生成单模型提交结果,包括lgb,xgb,ffm,mlp
  4. 打开./code/_4_5_model_avg.ipynb进行最终的加权平均并生成最终提交结果

方案说明

  1. 用户点击日志挖掘_2_1_gen_user_click_features.py
    挖掘广告点击日志,从不同时间粒度(天,小时)和不同属性维度(点击的素材,广告,推广计划,广告主类型,广告位等)提取用户点击行为的统计特征。
  2. 用户安装日志挖掘 _2_2_gen_app_install_features.py
    根据用户历史APP安装记录日志,分析用户的安装偏好和APP的流行趋势,结合APP安装时间的信息提取APP的时间维度的描述向量。这里最后只用了一种特征。
  3. 广告主转化回流上报机制分析_2_4_gen_tricks.py
    不同的广告主具有不同的转化计算方式,如第一次点击算转化,最后一次点击算转化,安装时点击算转化,分析并构造相应描述特征,提升模型预测精度。
  4. 广告转化率特征提取_2_5_gen_smooth_cvr.py
    构造转化率特征,使用全局和滑动窗口等方式计算单特征转化率,组合特征转化率,使用均值填充,层级填充,贝叶斯平滑,拉普拉斯平滑等方式对转化率进行修正。
  5. 广告描述向量特征提取_2_6_gen_ID_click_vectors.py
    广告投放是有特定受众对象的,而特定的受众对象也可以描述广告的相关特性,使用不同的人口属性对广告ID和APPID进行向量表示,学习隐含的语义特征。
  6. 建模预测
    使用多种模型进行训练,包括LightGBM,XGBoost,FFM和神经网络,最后进行多模型加权融合提高最终模型性能。

其他

  • 最终线上排名20,logloss 0.101763
  • 最终特征维度在110左右
  • 部分最终没有采用的特征代码依然保留
  • 由于我们团队的代码是3个人共同完成的,我这里整理的模型训练的部分可能和当时略有差异,但特征部分基本一致。
  • deprecated目录下为弃用的代码,包括一些原始代码和打算尝试的方法
Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].