Search code examples
pandaspycharmintellisense

Pandas dataframes and PyCharm IntelliSense

When I create new dataframes from old ones, using concat or merge, PyCharm intellisense stops working for the resulting dataframe unless I explicitly pass it to a DataFrame constructor

import pandas as pd

d1 = {1: [1, 2, 3], 2: [11, 22, 33]}
d2 = {1: [4], 2: [5]}

df1 = pd.DataFrame(d1)
df2 = pd.DataFrame(d2)
df3 = pd.concat([df1, df2], axis=0)
df3_ = pd.DataFrame(pd.concat([df1, df2], axis=0))

In the above example df3 and df3_ are the "same" dataframe, but intellisense only works on df3_. Am I doing something wrong? How can I avoid always having to call the DataFrame constructor and still get intellisense out of pycharm?

Solution

  • The answer is to use type hints like this:

    df3 = pd.concat([df1, df2], axis=0)  # type: pd.DataFrame