# numpy.diff() in Python

The numpy module of Python provides a function called numpy.diff for calculating the nth discrete difference along the given axis. If 'x' is the input array, then the first difference is given by out[i]=x[i+1]-a[i]. We can calculate the higher difference by using diff recursively. The numpy module of Python provides a function called numpy.diff for calculating the nth discrete difference along the given axis. If 'x' is the input array, then the first difference is given by out[i]=x[i+1]-a[i]. We can calculate the higher difference by using diff recursively.

### Syntax

snippet
`numpy.diff(a, n=1, axis=-1, prepend=<no value>, append=<no value>)`

### Parameters

x: array_like

This parameter defines the source array whose elements nth discrete deference are those which we want to calculate.

n: int(optional)

This parameter defines the number of times the values are differenced. If it is 0, then the source array is returned as it is.

append, prepend: array_like(optional)

This parameter defines a ndarray, which defines the values going to append or prepend to 'x', along the axis before computing differences.

### Returns:

This function returns a ndarray containing nth differences having the same shape as 'x,' and the dimension is smaller from n. The type of difference between any two elements of 'x' is the type of the output.

### Example 1:

snippet
```import numpy as np
arr = np.array([0, 1, 2], dtype=np.uint8)
arr
b=np.diff(arr)
b
arr[2,...] - arr[1,...] - arr[0,...]```

Output:

Output
array([0, 1, 2], dtype=uint8) array([1, 1], dtype=uint8) 1

In the above code

• We have imported numpy with alias name np.
• We have created an array 'arr' using np.array() function with the dtype 'uint8'.
• We have declared the variable 'b' and assigned the returned value of the np.diff() function.
• We have passed the array 'arr' in the function.
• Lastly, we tried to print the value of 'b' and the difference between elements.

In the output, it shows the discrete differences of elements.

### Example 2:

snippet
```import numpy as np
x = np.array([11, 21, 41, 71, 1, 12, 33, 2])
y = np.diff(x)
x
y```

Output:

Output
array([11, 21, 41, 71, 1, 12, 33, 2]) array([ 10, 20, 30, -70, 11, 21, -31])

### Example 3:

snippet
```import numpy as np
x = np.array([[11, 21, 41], [71, 1, 12], [33, 2, 13]])
y = np.diff(x, axis=0)
y
z = np.diff(x, axis=1)
z```

Output:

Output
array([[ 60, -20, -29], [-38, 1, 1]]) array([[ 10, 20], [-70, 11], [-31, 11]])

### Example 4:

snippet
```import numpy as np
x = np.arange('1997-10-01', '1997-12-16', dtype=np.datetime64)
y = np.diff(x)
y```

Output:

Output
array([1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1], dtype='timedelta64[D]')

In the above code

• We have imported numpy with alias name np.
• We have created an array of dates 'x' using np.arange() function with the dtype 'datetime64'.
• We have declared the variable 'y' and assigned the returned value of the np.diff() function.
• We have passed the array 'x' in the function.
• Lastly, we tried to print the value of 'y'.

In the output, it shows the discrete differences between dates.

Related Tutorial
Contents +