使用元组列表创建熊猫数据帧
原文:https://www . geesforgeks . org/creating-a-pandas-data frame-use-list-of-tuples/
我们可以从简单元组列表中创建一个 DataFrame,甚至可以选择我们想要使用的元组的特定元素。
代码#1: 简单地将元组传递给 DataFrame 构造函数。
# import pandas to use pandas DataFrame
import pandas as pd
# data in the form of list of tuples
data = [('Peter', 18, 7),
('Riff', 15, 6),
('John', 17, 8),
('Michel', 18, 7),
('Sheli', 17, 5) ]
# create DataFrame using data
df = pd.DataFrame(data, columns =['Name', 'Age', 'Score'])
print(df)
输出:
代码#2: 使用from_records()
import pandas as pd
# data in the form of list of tuples
data = [('Peter', 18, 7),
('Riff', 15, 6),
('John', 17, 8),
('Michel', 18, 7),
('Sheli', 17, 5) ]
# create DataFrame using data
df = pd.DataFrame.from_records(data, columns =['Team', 'Age', 'Score'])
print(df)
输出:
代码#3: 使用from_items()
import pandas as pd
# data in the form of list of tuples
data = [
('Age', [18, 15, 17, 18, 17]),
('Team', ['A', 'B', 'A', 'C', 'B']),
('Score', [7, 6, 8, 7, 5]),
]
# create DataFrame using data
df = pd.DataFrame.from_items(data)
print(df)
输出:
代码#4: 为枢转成为可能。
# import pandas to use pandas DataFrame
import pandas as pd
# data in the form of list of tuples
data = [('Peter', 18, 7),
('Riff', 15, 6),
('John', 17, 8),
('Michel', 18, 7),
('Sheli', 17, 5) ]
# create DataFrame using data
df = pd.DataFrame(data, columns =['Team', 'Age', 'Score'])
a = df.pivot('Team', 'Score','Age')
print(a)
输出:
版权属于:月萌API www.moonapi.com,转载请注明出处