CODEWITHSUNDEEP
pythonpandaslistdictionary
Crazy
Crazy

Reputation: 317

How to form a DataFrame from a list of dictionaries with list of values?

I have a list of list of dictionary with values in list that I would like to be transformed into a dataframe as shown below:

data_list = [
{'Width':[20,25,30,40],'Length':[21,22,23,24],'Ratio':[1.05,0.88,0.767,0.6]},
{'Width':[10,15,24,35],'Length':[20,25,30,40],'Ratio':[2,1.67,1.25,1.14]}
]

And I try to use pd.DataFrame.from_dict(data_list), however, the result is undesired as shown below:

|   |          Width |        Length |                 Ratio |
|---| -------------- | ------------  | --------------------- |
| 0 | [20,25,30,40]  | [21,22,23,24] | [1.05,0.88,0.767,0.6] |
| 1 | [10,15,24,35]  | [20,25,30,40] |    [2,1.67,1.25,1.14] |

What I expect is as below:

|   | Width | Length | Ratio |
|---| ----- | ------ | ----- |
| 0 |   20  |     21 |  1.05 |
| 1 |   25  |     22 |  0.88 |
......
| 6 |   24  |     30 |  1.25 |
| 7 |   35  |     40 |  1.14 |

Any help would be great! Thanks!

Upvotes: 3

Views: 87

Answers (3)

jezrael
jezrael

Reputation: 863741

Use solution from comments:

df = pd.DataFrame(data_list).apply(pd.Series.explode).reset_index(drop=True)
print (df)
  Width Length  Ratio
0    20     21   1.05
1    25     22   0.88
2    30     23  0.767
3    40     24    0.6
4    10     20      2
5    15     25   1.67
6    24     30   1.25
7    35     40   1.14

Another idea is use defaultdict with extend, last pass to DataFrame constructor for improve performance:

from collections import defaultdict
d = defaultdict(list)

for di in data_list:
    for k, v in di.items():
        d[k].extend(v)
        
df = pd.DataFrame(d)
print (df)
   Width  Length  Ratio
0     20      21  1.050
1     25      22  0.880
2     30      23  0.767
3     40      24  0.600
4     10      20  2.000
5     15      25  1.670
6     24      30  1.250
7     35      40  1.140

Upvotes: 1

sammywemmy
sammywemmy

Reputation: 28729

Another option would be to concatenate a list of DataFrames:

pd.concat([pd.DataFrame(ent) for ent in data_list], ignore_index = True)

   Width  Length  Ratio
0     20      21  1.050
1     25      22  0.880
2     30      23  0.767
3     40      24  0.600
4     10      20  2.000
5     15      25  1.670
6     24      30  1.250
7     35      40  1.140

Upvotes: 0

TaekYoung
TaekYoung

Reputation: 110

I think this can do your need if you don't need index.

import pandas as pd


data_list = [
{'Width':[20,25,30,40],'Length':[21,22,23,24],'Ratio':[1.05,0.88,0.767,0.6]},
{'Width':[10,15,24,35],'Length':[20,25,30,40],'Ratio':[2,1.67,1.25,1.14]}
]

for data in data_list:
    df = df.append(pd.DataFrame(data))

Upvotes: 0

Related Questions