Numpy Basics (I)
Array properties
The dimension of NumPy array is called rank. Rank is the number of axes, that is, the dimension of the array.
In NumPy, each linear array is called an axis, that is, dimensions. For example, a two-dimensional array is equivalent to two one-dimensional arrays, in which each element in the first one-dimensional array is a one-dimensional array. Therefore, a one-dimensional array is the axis in NumPy. The first axis is equivalent to the underlying array, and the second axis is the array in the underlying array. The number of axes, the rank, is the dimension of the array.
Many times you can declare axis. axis=0, indicating operation along axis 0, i.e. operation on each column; axis=1 indicates the operation along the first axis, that is, the operation is performed on each line.
| attribute | explain |
|---|---|
| ndarray.ndim | Rank is the number of axes or dimensions |
| ndarray.shape | The dimension of the array. For a matrix, n rows and m columns |
| ndarray.size | The total number of array elements, equivalent to Value of n*m in shape |
| ndarray.dtype | The element type of the ndarray object |
| ndarray.itemsize | The size, in bytes, of each element in the ndarray object |
| ndarray.flags | Memory information of the ndarray object |
| ndarray.real | The real part of the ndarray element |
| ndarray.imag | Imaginary part of ndarray element |
| ndarray.data | The buffer containing the actual array elements usually does not need to use this attribute because the elements are generally obtained through the index of the array. |
Create array
#uninitialized numpy.empty(shape, dtype = float, order = 'C') #Initialize to 0 numpy.zeros(shape, dtype = float, order = 'C') #Initialize to 1 numpy.ones(shape, dtype = None, order = 'C')
| parameter | describe |
|---|---|
| shape | Array shape |
| dtype | Data type, optional |
| order | There are "C" and "F" options, representing row priority and column priority respectively, and the order of storage elements in computer memory. |
Create an array from a numeric range
numpy.arange(start, stop, step, dtype) #Start value, end value, step size, returned data type #Generate an array of 0 ~ 5 x = np.arrange(5) #numpy. The linspace function is used to create a one-dimensional array, which is composed of an arithmetic sequence np.linspace(start, stop, num=50, endpoint=True, retstep=False, dtype=None) #num number of samples not generated (equal step size guaranteed) endpoint is fasle, and stop is not included in the sequence #When retstep is true, the spacing will be displayed in the generated array #numpy. The logspace function is used to create an equal ratio sequence np.logspace(start, stop, num=50, endpoint=True, base=10.0, dtype=None) #Base is the base of log a = np.logspace(0,9,10,base=2) print (a) #[ 1. 2. 4. 8. 16. 32. 64. 128. 256. 512.]
Indexes
a = np.arange(10) s = slice(2, 7, 2) #From index 2 to index 7, the step is 2 ic(a[s]) ic(a[2:7:2])#Three parameters start, stop and step can be separated by colons ic(a[2:])#Represents all elements after 2
x = np.array([[1, 2], [3, 4], [5, 6]]) #Gets the element at position (0,0) (1,1) (2,0) y = x[[0, 1, 2], [0, 1, 0]] #Get results [1, 4, 5]
x = np.array([[0, 1, 2], [3, 4, 5], [6, 7, 8], [9, 10, 11]])
print("Array is:", x)
print('\n')
# index = np.array([0, 0, 3, 3], [0, 2, 0, 2]) cannot be written like this
ic(x[[0, 0, 3, 3], [0, 2, 0, 2]]) #Generate a list
rows = np.array([[0, 0], [3, 3]])
cols = np.array([[0, 2], [0, 2]])
ic(x[rows, cols]) #Generate a matrix with 2 rows and 2 columns
#result
x[[0, 0, 3, 3], [0, 2, 0, 2]]: array([ 0, 2, 9, 11])
x[rows, cols]: array([[ 0, 2],
[ 9, 11]])
a = np.array([[1,2,3], [4,5,6],[7,8,9]]) ic(a[1:3, 1:3]) ic(a[1:3, [1, 2]]) #Note that 1:3 has the same function as [1, 2] ic(a[:, 1:]) ic(a[..., 1:]) #... And: similar in function #The results are as follows: #ic| a[1:3, 1:3]: array([[5, 6], # [8, 9]]) #ic| a[1:3, [1, 2]]: array([[5, 6], # [8, 9]]) #ic| a[:, 1:]: array([[2, 3], # [5, 6], # [8, 9]]) #ic| a[..., 1:]: array([[2, 3], # [5, 6], # [8, 9]])
Boolean index (filter to get a specific value)
x = np.array([[0, 1, 2], [3, 4, 5], [6, 7, 8], [9, 10, 11]]) ic(x) ic(x[x > 5]) #ic| x: array([[ 0, 1, 2], # [ 3, 4, 5], # [ 6, 7, 8], # [ 9, 10, 11]]) #ic| x[x > 5]: array([ 6, 7, 8, 9, 10, 11])
#Using negation operator~ a = np.array([np.nan, 1, 2, np.nan, 3, 4, 5]) ic(a[~np.isnan(a)]) #array([1., 2., 3., 4., 5.])
Fancy index
The integer array is used for indexing, and each value is taken as the following table of an axis of the target array,
x = np.arange(32).reshape((8, 4)) y = np.arange(8) index = [4, 3, 2, 1] ic(x) ic(y) ic(x[index])#A one-dimensional index of a two-dimensional array returns the contents of each row ic(y[index])#The one-dimensional index of the one-dimensional array returns the corresponding element #x[index]: array([[16, 17, 18, 19], # [12, 13, 14, 15], # [ 8, 9, 10, 11], # [ 4, 5, 6, 7]]) #y[index]: array([4, 3, 2, 1])
NP. Is required to pass in multiple index arrays ix_
x=np.arange(32).reshape((8,4)) print (x[np.ix_([1,5,7,2],[0,3,1,2])]) #[[ 4 7 5 6] #[20 23 21 22] #[28 31 29 30] #[ 8 11 9 10]]