numpy 自己生成数据集

一、random

1. seed(int n) 随机数种子（其实生成随机数是根据随机数种子使用一定的算法生成随机数的，因此也可以说是伪随机数）

2. rand(int n) 产生n个均匀分布的样本值

import numpy as np
from numpy import random
print(random.rand(10))
"""
输出：
[0.00279392 0.30422703 0.40830302 0.76097027 0.88393813 0.95769368
 0.54477196 0.91185166 0.91407798 0.54808962]
"""

3. randint(begin,end,num) begin和end之间随机生成num个数

import numpy as np
from numpy import random
print(random.randint(10,20,5))·
"""
输出：
[15 10 10 10 16]
"""

4. randn(N,M) 生成N*M的符合正态分布（均值为0，标准差为1）的数组

import numpy as np
from numpy import random
print(random.randn(10,5))
"""
[[-0.14930229 -0.49723555  0.05889083 -1.18325212  0.09198058]
 [-0.12299026 -0.50251851 -1.11138151 -0.14130106  1.96066248]
 [-0.86989821  1.55537731  0.14948361  0.65432214 -1.18670275]
 [ 0.22100948 -0.61138931  0.61774483  0.19456671 -2.07787105]
 [-0.20535281  0.08024417  0.13413211 -1.56538371  0.88922951]
 [-0.79801568 -0.41154799 -0.91625006  1.44174114  1.87717894]
 [ 1.31754142  0.19543874 -1.17530768  0.36376645 -0.07837382]
 [ 0.89909056  0.48339419  1.32612804 -0.22504098 -0.31849442]
 [-0.57056038  0.13980028  1.116261    1.27374768 -0.25722175]
 [-1.13907277 -0.3458683   0.10948824 -0.45425931 -0.30000255]]
"""

5. normal(N,M) 生成N*M的符合普通正态分布的数组

import numpy as np
from numpy import random
print(random.normal(size=(10,5)))
"""
[[ 1.07713963  0.90726022 -0.42696564 -1.47677281 -0.07285344]
 [-0.85633398  1.88888854  0.48570094  0.23756726  0.48340864]
 [-0.31309425 -0.65975846  0.31369926 -0.52956323  1.43464665]
 [-1.45134155  0.29923932 -0.69863863 -0.27948516 -1.21324693]
 [-0.11547236  1.12712333 -0.67411053 -0.36735838  0.27807234]
 [ 1.59508896 -1.59286109  0.87856752  1.67809503  0.27672422]
 [ 0.61031653 -0.40900192  0.1169569  -0.07748274 -2.32155114]
 [-0.21815467  0.0928035  -0.94198498  2.60294661 -0.75585361]
 [-0.39176799  0.42874459 -0.92383274  0.44186166  0.19437499]
 [-0.77975909  0.11364601 -0.23020196  0.28517307  1.09343548]]
"""

6. uniform(min,max) min到max中随机生成一个数

import numpy as np
from numpy import random
print(random.uniform(10,100))

7. Beta(ndarray1,ndarray2) 生成beta分布样本值，参数大于0
8. gamma（ndarray1,ndarray2） 生成gamma分布样本值

二、linspace

numpy.linspace(start,end,num,endpoint=True,retstep=False,dtype=None)
start ：开始位置
end ：结束位置
num ：数据个数
endpoint ：是否包含结束位置（默认是True,包含）
retstep ：是否显示间距（如果显示的还，第一个元素data[0]显示数组，第二个元素data[1]显示间距）
dtype ：数据类型

import numpy as np
lin_data = np.linspace(10,100,15)
print(lin_data)
"""
[ 10.          16.42857143  22.85714286  29.28571429  35.71428571
  42.14285714  48.57142857  55.          61.42857143  67.85714286
  74.28571429  80.71428571  87.14285714  93.57142857 100.        ]
"""

import numpy as np
lin_data = np.linspace(10,100,15,retstep=True)
print(lin_data)
"""
(array([ 10.        ,  16.42857143,  22.85714286,  29.28571429,
        35.71428571,  42.14285714,  48.57142857,  55.        ,
        61.42857143,  67.85714286,  74.28571429,  80.71428571,
        87.14285714,  93.57142857, 100.        ]), 6.428571428571429)
"""

lin_data[0]
"""
array([ 10.        ,  16.42857143,  22.85714286,  29.28571429,
        35.71428571,  42.14285714,  48.57142857,  55.        ,
        61.42857143,  67.85714286,  74.28571429,  80.71428571,
        87.14285714,  93.57142857, 100.        ])
"""
lin_data[1]
"""
6.428571428571429
"""