You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
emmm目测当时想减少内存占用,计算出来大概都是四五百,相当于去掉百位,459>-59,应该有用吧..------------------ 原始邮件 ------------------
发件人: "peterpanhjh123"<[email protected]>
发送时间: 2020年1月10日(星期五) 中午11:21
收件人: "liupengsay/2018-Tencent-social-advertising-algorithm-contest"<2018-Tencent-social-advertising-algorithm-contest@noreply.github.com>;
抄送: "Subscribed"<[email protected]>;
主题: [liupengsay/2018-Tencent-social-advertising-algorithm-contest] 关于转化率特征部分的问题 (#2)
作者您好,想请问下在构造转化率特征的过程中,为什么要-400?
如下代码
df_feature[col_name] = ((pd.Series(data.index).map(se)*10000)-400).fillna(value=-1).astype('int')
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or unsubscribe.
作者您好,想请问下在构造转化率特征的过程中,为什么要-400?
如下代码
df_feature[col_name] = ((pd.Series(data.index).map(se)*10000)-400).fillna(value=-1).astype('int')
The text was updated successfully, but these errors were encountered: