shenweichen / Tencent_social_ads2017_mobile_app_pcvr
Tencent Social Ads 2017 contest rank 20
Stars: ✭ 155
Labels
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
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
- Keras==2.0.6
- lightgbm==0.1
- matplotlib==2.0.0
- numpy==1.11.3
- pandas==0.19.2
- scikit-learn==0.18.1
- scipy==0.18.1
- tensorflow-gpu==1.2.1
- tqdm==4.11.2
- xgboost==0.6a2
- 其他库 LIBFFM v121
运行说明
- 将复赛数据文件
final.zip
放在根目录下 - 在根目录下运行
sh run.sh
命令生成特征文件 - 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
- 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
- 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。 - 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。 - 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。 - 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。 - 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。 - 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].