Need to convert an Array of Dicts (complex) into a Pandas Dataframe

Question

This is my structure :

data = [{'dev_name': 'some_name',
'dev_connects': ['cn1', 'cn2', 'cn3', 'cn4'], 
'dev_type': 'some_type', 
'dev_features': ['ft 1', 'ft 2']},

{'dev_name': 'some_name2', 
'dev_connects': ['cn1', 'cn2', 'cn3', 'cn4'], 
'dev_type': 'some_type2', 
'dev_features': ['ft 1', 'ft 2']}]

So as you can see from my description , its an array of dicts. I need to be able to create a dataframe that will read each object and columnize every key, including what is inside the arrays data[i]['dev_connects'] and data[i]['dev_features'] for every object :

dev_name	connect1	connect2	connect3	connect4	dev_type	feature1	feature2
some_name	cn1	cn2	cn3	cn4	some_type	ft1	ft2
some_name2	cn1	cn2	cn3	cn4	some_type2	ft1	ft2

Does this answer your question? Split a Pandas column of lists into multiple columns — Ynjxsjmh
– Ynjxsjmh, Commented Aug 5, 2022 at 7:14

mattdonders · Accepted Answer · 2022-08-05 06:32:23Z

I was able to accomplish this by breaking up your DataFrames into two separate (via apply(pd.Series) and then merging them back together. See the final code below.

df_base = pd.DataFrame(data)[["dev_name", "dev_type"]]
df_connects = pd.DataFrame(data).set_index("dev_name")["dev_connects"].apply(pd.Series).add_prefix("connect")
df_features = pd.DataFrame(data).set_index("dev_name")["dev_features"].apply(pd.Series).add_prefix("feature")

df_final = pd.merge(df_base, df_connects, on="dev_name").merge(df_features, on="dev_name")

This leaves you with the following final result.

	dev_name	dev_type	connect0	connect1	connect2	connect3	feature0	feature1
0	some_name	some_type	cn1	cn2	cn3	cn4	ft 1	ft 2
1	some_name2	some_type2	cn1	cn2	cn3	cn4	ft 1	ft 2

Collectives™ on Stack Overflow

Need to convert an Array of Dicts (complex) into a Pandas Dataframe

1 Answer 1

Comments

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Linked

Related