+- Python Forum (https://python-forum.io)
+-- Forum: Python Coding (https://python-forum.io/forum-7.html)
+--- Forum: Data Science (https://python-forum.io/forum-44.html)
+--- Thread: Pandas nested json data to dataframe (/thread-12183.html)
Pandas nested json data to dataframe - FrankC - Aug-13-2018
We have this json response:
https://api.binance.com/api/v1/exchangeInfo
And want to get each symbol filters content in a dataframe.
How can we do it?
RE: Pandas nested json data to dataframe - scidam - Aug-14-2018
To interpret the json-data as a DataFrame object Pandas requires the same length
of all entries. So,
pd.read_json(...) will fail to convert data to a valid DataFrame. However, you can load it as a Series, e.g.
import pandas as pd
data = pd.read_json('https://api.binance.com/api/v1/exchangeInfo', typ='series')From now, you can follow different ways to interpret the data as a DataFrame object: use Pandas MultiIndex and build a panel data structure, or build a DataFrame object directly, e.g.my_df = pd.concat([pd.DataFrame(data.symbols[j]['filters']).assign(symbol_id=j) for j in range(len(data.symbols))]).reset_index()Note, in this code I used
assign method to store symbol ids (it might be useful for future analysis).