Python - Pandas - 扁平化字典列列表

Question

I've imported a.csv as dataframe but I have a column composed as list of dicts. How can I flatten it into multiple columns/new df?

example below:

Answer 1

You can extract each row into a series and concat them to a dataframe:

import pandas as pd

data = []
for row in range(len(df_ass)):
    data.append(pd.Series(df_ass['associations'][row]['order_rows'][0]))
    
df = pd.concat(data)

Answer 2

With a dataframe like

df = pd.DataFrame({
    "associations" :[
        '{"order_rows": [{"id": 1, "p_id": 12}, {"id": 2, "p_id": 12}]}',
        '{"order_rows": [{"id": 3, "p_id": 13}, {"id": 4, "p_id": 14}, {"id": 5, "p_id": 15}]}'
    ]
})

you can do

df = pd.DataFrame(df.associations.map(eval).str.get("order_rows").explode().to_list())

or

df = pd.json_normalize(df.associations.map(eval), record_path="order_rows")

to get

   id  p_id
0   1    12
1   2    12
2   3    13
3   4    14
4   5    15

Answer 3

solved like this:

df=pd.read_csv(fName, sep=",", usecols=["associations"]) 
list_rows_dicts=[]
for index, row in df.iterrows():
    row_dict=eval(row["associations"])
    for d in row_dict["order_rows"]:
        list_rows_dicts.append(d)
csv_final_df=pd.DataFrame(list_rows_dicts)