Question

我想合并从给定函数创建的多个数据框。我尝试使用pd.concat，但收到错误消息：

TypeError：第一个参数必须是pandas对象的可迭代对象，您传递了一个类型为“ DataFrame”的对象

这是我编写的代码：

def plot_percs(sims,name):
    filter = name
    session_to_leads = (sims['session_to_leads'].sum()/len(sims['session_to_leads']))*100
    leads_to_opps = (sims['lead_to_opps'].sum()/len(sims['lead_to_opps']))*100
    oops_to_comp = (sims['oops_to_comp'].sum()/len(sims['oops_to_comp']))*100
    percentages = pd.DataFrame(data=[filter,session_to_leads,leads_to_opps,oops_to_comp]).transpose()
    percentages.columns = ['filter','session_to_leads','leads_to_opps','opps_to_complete']
    return percentages
pd.concat(plot_percs(originalsims,'original'),plot_percs(facebooksims,'facebook'))

这是两个数据框的示例 originalsims：

    filter  session_to_leads    leads_to_opps   opps_to_complete
 0 original       16.28               9.47           1.92

facebooksims

    filter  session_to_leads    leads_to_opps   opps_to_complete
0  facebook       1.76                0.02             0

Answer 1

为传递给[]函数的DataFrame的列表添加pd.concat，也可以添加ignore_index=True以避免在输出中出现重复的索引：

df = pd.concat([plot_percs(originalsims,'original'),
                plot_percs(facebooksims,'facebook')], ignore_index=True)

我认为您的函数应该用mean重写：

def plot_percs(sims,name):
    c = ['session_to_leads', 'lead_to_opps', 'oops_to_comp']
    d = {'oops_to_comp':'opps_to_complete'}
    percentages = sims[c].mean().to_frame().T.rename(columns=d)
    percentages.insert(0, 'filter', name)
    return percentages

使用pd.concat合并多个数据帧

1 个答案: