Machine learning -- pandas is easy to use

Pandas

Data Processing Tools in Machine Learning

In [1]:

# import pandas

import pandas as pd

print (pd.__version__)

0.22.0

main data structure

• DateFrame data frame, which can be considered as a two-dimensional data table
• Series data sequence, which can be considered one-dimensional

In [2]:

# Series

s = pd.Series(['a', 'b', 'c'])

print(s)

0    a
1    b
2    c
dtype: object

In [7]:

# DataFrame

score = pd.Series([2,6,2])

df = pd.DataFrame({"s_name": s, "s_score": score})

print (df)

# df.describe()

df.head(2)

  s_name  s_score
0      a        2
1      b        6
2      c        2

Out[7]:

	s_name	s_score
0	a	2
1	b	6

In [10]:

# pandas plot

import matplotlib.pyplot as plt

cc = pd.read_csv("california_housing_train.csv", sep=',')

cc.hist("housing_median_age")

​

Out[10]:

array([[<matplotlib.axes._subplots.AxesSubplot object at 0x0000000009617B38>]],
      dtype=object)

In [14]:

score/1.5

Out[14]:

0    1.333333
1    4.000000
2    1.333333
dtype: float64

In [25]:

import numpy as np

s.index

s.reindex(np.random.permutation(s.index))

Out[25]:

2    c
0    a
1    b
dtype: object

In [24]:

s.reindex(np.random.permutation(4))

Out[24]:

0      a
2      c
1      b
3 NaN
dtype: object

In [ ]:

​