TensorFlow can solve a series of linear equations using the solve operation. Let's first explain this without using the library and later use the solve function.
A linear equation is represented as follows:
ax + b = yy - ax = b
y - ax = b
y/b - a/b(x) = 1
Our job is to find the values for a and b in the preceding equation, given our observed points. First, create the matrix points. The first column represents x values, while the second column represents y values.
Consider that X is the input matrix and A is the parameters that we need to learn; we set up a system like AX=B, therefore, 
. The following example, with code, ...
