pandas如何编写查询

Question

基于此 pandas get column average/mean

我可以像这样创建一个简单的计算字段： 我的查询

df = pd.read_sql("select range_start, range_end  from "+table+" group by  range_start, range_end", conn)

创建此表：

Start   Stop
4385159 4499467
4175786 4352309
342426  354137
5591040 5600392

我想要做的是注入一个具有差异的列，我可以这样做：

df2['Diff'] = df2['Stop'] - df2['Start']

现在我的表看起来像这样：

Start   End      Diff
4385159 4499467 114308
4175786 4352309 176523
342426  354137  11711

我的问题是如何编写将返回结果的查询：

df = pd.read_sql("select Diff  from "+table+" where Diff < Xnumber group by  Diff", conn)

我想我需要在jupyter（pandas）中查询查询。做这样的事情：

df = pd.read_sql("select (df2['Stop'] - df2['Start']) as df2['Diff'] where (df2['Stop'] - df2['Start']) < Xnumber group by (df2['Stop'] - df2['Start'])",conn)

^没有用但你明白了

Answer 1

我可能会遗漏一些东西，但您是否可以直接在pandas中创建新列，而不进行任何查询？

df['Diff'] = df2['Stop'] - df2['Start']

Answer 2

知道了：

df6 = pd.read_sql("select (Start - Stop) as Diff from "+table+" where <condition>",conn)