Numpy 或 Pandas 删除 全零行、全零列
使用numpy、pandas 实现删除全0的行或者列。
pandas删除全零行
from pandas import DataFrame
import numpy as np
df1 = DataFrame(np.arange(16).reshape((4,4)),index=['a','b','c','d'],columns=['one','two','three','four']) # 创建一个dataframe
df1.loc['e'] = 0 # 优雅地增加一行全0
print(df1)
df1.ix[(df1==0).all(axis=1), :] # 找到它
df2 = df1.ix[~(df1==0).all(axis=1), :] # 删了它
print("======+=========")
print(df2)
'''
one two three four
a 0 1 2 3
b 4 5 6 7
c 8 9 10 11
d 12 13 14 15
e 0 0 0 0
======+=========
one two three four
a 0 1 2 3
b 4 5 6 7
c 8 9 10 11
d 12 13 14 15
'''
pandas删除全零列
from pandas import DataFrame
import numpy as np
df1 = DataFrame(np.arange(16).reshape((4,4)),index=['a','b','c','d'],columns=['one','two','three','four']) # 创建一个dataframe
df1.loc[:,['four']] = 0 # 将一列全部设置为0
print(df1)
df1.ix[:, (df1==0).all(axis=0)] # 找到它
df2 = df1.ix[:, ~(df1==0).all(axis=0)] # 删了它
print("======+=========")
print(df2)
'''
one two three four
a 0 1 2 0
b 4 5 6 0
c 8 9 10 0
d 12 13 14 0
======+=========
one two three
a 0 1 2
b 4 5 6
c 8 9 10
d 12 13 14
'''
numpy 删除全零行
import numpy as np
# 生成50个[0,10]间的随机数
data = np.random.choice(10, 50)
# 变形
data = data.reshape(10,5)
# 将这三行全设为0
data[3,:] = 0
data[7,:] = 0
data[8,:] = 0
print(data)
print('original data shape', data.shape)
print('----- after removing rows with only 0 -----')
# 删除全零行
# To remove all rows that contain only 0 we can use the following syntax
data = data[~np.all(data == 0, axis=1)]
print(data)
print('new data shape', data.shape)
'''
[[8 6 2 2 8]
[5 3 0 2 5]
[3 2 0 3 8]
[0 0 0 0 0]
[9 5 4 8 3]
[9 7 8 6 4]
[0 8 2 0 7]
[0 0 0 0 0]
[0 0 0 0 0]
[5 7 3 7 0]]
original data shape (10, 5)
----- after removing rows with only 0 -----
[[8 6 2 2 8]
[5 3 0 2 5]
[3 2 0 3 8]
[9 5 4 8 3]
[9 7 8 6 4]
[0 8 2 0 7]
[5 7 3 7 0]]
new data shape (7, 5)
'''numpy 删除全零列
示例1
关键代码：
# 寻找并删除全零列
data= np.delete(data,np.where(~data.any(axis=0))[0], axis=1)import numpy as np
# 生成50个[0,10]间的随机数
data = np.random.choice(10, 50)
# 变形
data = data.reshape(5,10)
# 将这三列全设为0
data[:,2] = 0
data[:,4] = 0
data[:,7] = 0
print(data)
print('original data shape', data.shape)
print('----- after removing columns with only 0 -----')
# 寻找全零列
print(np.where(~data.any(axis=0))[0])
# 删除全零列
# To remove all columns that contain only 0 we can also use the following syntax
data= np.delete(data,np.where(~data.any(axis=0))[0], axis=1)
print(data)
print('new data shape', data.shape)
'''
[[6 3 0 0 0 0 1 0 6 7]
[2 8 0 2 0 6 2 0 2 5]
[6 8 0 9 0 1 3 0 0 0]
[3 1 0 6 0 3 7 0 2 5]
[8 5 0 2 0 4 9 0 8 2]]
original data shape (5, 10)
----- after removing columns with only 0 -----
[2 4 7]
[[6 3 0 0 1 6 7]
[2 8 2 6 2 2 5]
[6 8 9 1 3 0 0]
[3 1 6 3 7 2 5]
[8 5 2 4 9 8 2]]
new data shape (5, 7)
'''
示例2
关键代码：
# 删除全零列
mask = (array1 == 0).all(0)
column_indices = np.where(mask)[0]
array1 = array1[:,~mask]import numpy as np
# 生成50个[0,10]间的随机数
array1 = np.random.choice(10, 50)
# 变形
array1 = array1.reshape(5,10)
# 将某些列设为0
array1[:,2]=0
array1[:,5]=0
array1[:,8]=0
print(array1)
print("raw array", array1.shape)
# 删除全零列
mask = (array1 == 0).all(0)
column_indices = np.where(mask)[0]
array1 = array1[:,~mask]
print("================")
print("after array",array1.shape)
print(array1)
'''
[[1 6 0 9 9 0 4 2 0 0]
[8 1 0 3 6 0 1 9 0 6]
[2 5 0 7 2 0 8 5 0 5]
[8 2 0 3 2 0 9 1 0 6]
[3 9 0 5 2 0 3 8 0 2]]
raw array (5, 10)
================
after array (5, 7)
[[1 6 9 9 4 2 0]
[8 1 3 6 1 9 6]
[2 5 7 2 8 5 5]
[8 2 3 2 9 1 6]
[3 9 5 2 3 8 2]]
'''
示例3
关键代码：
# 删除全零列
idx = np.argwhere(np.all(array1[..., :] == 0, axis=0))
array1= np.delete(array1, idx, axis=1)import numpy as np
# 生成50个[0,10]间的随机数
array1 = np.random.choice(10, 50)
# 变形
array1 = array1.reshape(5,10)
# 将某些列设为0
array1[:,2]=0
array1[:,5]=0
array1[:,8]=0
print(array1)
print("raw array", array1.shape)
# 删除全零列
idx = np.argwhere(np.all(array1[..., :] == 0, axis=0))
array1= np.delete(array1, idx, axis=1)
print("================")
print("after array",array1.shape)
print(array1)
'''
[[7 8 0 8 0 0 0 4 0 2]
[9 4 0 2 4 0 2 4 0 1]
[4 7 0 5 5 0 3 3 0 3]
[5 0 0 9 8 0 3 2 0 7]
[3 4 0 3 2 0 5 0 0 6]]
raw array (5, 10)
================
after array (5, 7)
[[7 8 8 0 0 4 2]
[9 4 2 4 2 4 1]
[4 7 5 5 3 3 3]
[5 0 9 8 3 2 7]
[3 4 3 2 5 0 6]]
'''
参考博客
How to remove array rows that contain only 0 in python (moonbooks.org)
Numpy（Pandas） 删除全为零的列_SongpingWang的博客-CSDN博客_pandas删除全为0的列