You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
|
|
|
|
### 功能说明
|
|
|
|
|
基于pytorch的Deep&Cross实现广告点击预测
|
|
|
|
|
### 使用说明
|
|
|
|
|
1. 下载:
|
|
|
|
|
使用`git clone https://bdgit.educoder.net/pseuc2htk/DeepCross.git`命令下载,或者直接下载zip包解压使用。
|
|
|
|
|
|
|
|
|
|
2. 环境配置:
|
|
|
|
|
要求python>=3.6,并配置好GPU环境,其他环境配置如下:
|
|
|
|
|
> pytorch>=1.4
|
|
|
|
|
cuda=10.0
|
|
|
|
|
cuDNN =7.6
|
|
|
|
|
sklearn
|
|
|
|
|
numpy
|
|
|
|
|
pandas
|
|
|
|
|
|
|
|
|
|
3. 路径设置
|
|
|
|
|
设置`./tool/train_test.py`文件下路径,保证能正常访问数据。
|
|
|
|
|
|
|
|
|
|
4. 模型训练和推理
|
|
|
|
|
在`./tool`文件夹下,配置`./tool/train_test.py`下训练参数后,运行:
|
|
|
|
|
`python train_test.py`
|
|
|
|
|
进行训练和推理。
|
|
|
|
|
|
|
|
|
|
5. 运行`train_test.py`中的推理部分,对给出的测试数据进行预测并打印出预测值。
|
|
|
|
|
|
|
|
|
|
6. 将打印出的预测值粘贴到头歌测试平台该关卡下右侧代码编辑框内`(result/result.txt)`进行评测;
|
|
|
|
|
|
|
|
|
|
|