Tensors, or tensors, can be seen as a natural generalization of vectors and matrices, and we use tensors to represent a wide range of data types.
The order of a tensor is sometimes called a dimension, or axis, and the word axis is translated from the English axis.
For example, a matrix [[1,2],[3,4]] is a 2-order tensor with two dimensions or axes.
What you see along the 0th axis is [1,2], [3,4] two vectors, which are equivalent to a 2x2 matrix in mathematics, taken out by row, each row as a vector;
What you see along the first axis is [1,3], [2,4] two vectors, equivalent to a 2x2 matrix in mathematics, taken out by columns, each column as a vector;
import numpy as np a = np.array([[1,2],[3,4]]) s0 = np.sum(a, axis=0) #ie [1,2], [3,4] addition s1 = np.sum(a, axis=1) #ie [1,3], [2,4] addition print(s0) print(s1)
The result is displayed as:
>>> [4, 6]
>>> [3, 7]