This is the 24th day of my participation in the August More Text Challenge

1. Type conversion

Different types of values can perform different operations, so cast the raw data. In numpy arrays, the conversion method is astype(), which specifies the target type to be converted in parentheses after astype.

import numpy as np
​
arr = np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]])
print(arr)
Copy the code

result:

[1 2 3] [4 5 6] [7 8 9]]Copy the code
print(arr.dtype)
Copy the code

result:

int32
Copy the code

Convert an ARR array from int to float

print(arr.astype(np.float64))
Copy the code

result:

[[1. 2. 3.] [4. 5. 6.] [7.Copy the code

The astype() methods here and in the Pandas series are essentially the same, although they belong to two different libraries.

2. Handle the missing value

Missing values are represented by Np.nan in numpy

Missing value processing is divided into 2 steps. The first step is to judge whether there are missing values and find out the missing values. The second step is to fill in the missing values

An array of missing values is given before processing the missing values

import numpy as np
​
arr = np.array([1, 2, np.nan, 4])
print(arr)
Copy the code

result:

[ 1.  2. nan  4.]
Copy the code

2.1 Searching for missing Values

The method used to find missing values is isnan(). If a location returns True, it is a missing value.

print(np.isnan(arr))
Copy the code

result:

[False False  True False]
Copy the code

2.2 Handling missing Values

arr[np.isnan(arr)] = 0
print(arr)
Copy the code

result:

/ 1. 2. 0. 4.Copy the code

3. Repeat value processing

Start by creating an array of duplicate values

import numpy as np
​
arr = np.array([1, 2, 3, 2, 1])
print(arr)
Copy the code

result:

[1 2 3 2 1]
Copy the code
print(np.unique(arr))
Copy the code

result:

[1, 2, 3]Copy the code