收录一写深度学习相关的python代码的使用心得
1 写在开头
记录一些好代码的使用方法
2 DataFrame
默认导入的设置为:
import numpy as np
2.1 pd.date_range
Use to: 用来生成时间间隔的代码,效率很高。
Return type: DatatimeIndex
Demo:
startTime = '2017-03-01 00:00:00'
endTime = '2017-05-31 23:58:00'
fullTimeRange = pd.date_range(pd.to_datetime(startTime), pd.to_datetime(endTime), freq='2min')
2.2 df.unique()
Use to: 从结果可以看到unique()相当于把list变为set
Return type: ndarray
Demo:
Input of demo:
test = pd.Series([1,1,2,2,3,3,3,4,4])
res = test.unique()
print(res)
Output of demo:
[1, 2, 3, 4]
2.3 pd.MultiIndex.from_product
Use to: Make a MultiIndex from the cartesian product of multiple iterables.
Return type: DataFrame
Demo:
Input of demo:
numbers = [1,2,3]
colors = ['green','purple']
pd.MultiIndex.from_product([numbers, colors], names=['number','color'])
Output of demo:
MultiIndex([(0, 'green'),
(0, 'purples'),
(1, 'green'),
(1, 'purples'),
(2, 'green'),
(2, 'purples')],
names=['number', 'color'])
2.4 df.groupby()
Use to: 按照特定的方式将列表重新划分,模型按照列划分。
Return type: DataFrame
Demo
2.5 df.pivot()
Use to: 将表格转化为透视表,在实际问题中很有用
Return type: DataFrame