Learn to use Symbolics.jl

General Usage
,
1

Hi , I want to use Symbolics.jl to solve bo and b1 from 2 equations
b0 + b1 * x1 = y1
b0 + b1 * x2 = y2

I know in SymPy.jl it would be

using SymPy
@syms x[1:2] y[1:2] b[0:1]
eqns = [one.(x) x] * b .~ y
solve(eqns, b)

now i can build this

using Symbolics, Groebner, Nemo, LinearAlgebra
@variables x[1:2] y[1:2] b b0
eqs = [b0 + b * x[1] ~ y[1] 
       b0 + b * x[2] ~ y[2] ]
symbolic_solve(eqs, [b, b0])

Is it possible to build it with only 1 equation eqns = [one.(x) x] * b .~ y as in SymPy.jl ?

I tried

@variables x[1:2] y[1:2] b[0:1]
eqs = [b[0] + b[1] * x[1] ~ y[1] 
       b[0] + b[1] * x[2] ~ y[2] ]
symbolic_solve(eqs, b)

Warning: Solve can not solve this input currently

@variables x[1:2] y[1:2] b[0:1]
eqs = [ones(length(x)) x] * b .~ y

ERROR: DimensionMismatch: expected axes(b, 1) = Base.OneTo(2)

b = Symbolics.variables(:b, 0:1)
x = Symbolics.variables(:x, 1:2)
y = Symbolics.variables(:y, 1:2)

eqs = [[ones(length(x)) x] * b .~ y]
symbolic_solve(eqs, b)

ERROR: AssertionError: Invalid input

@variables x, y, b
p = [ones(length(x)) x] * b .~ y
f = build_function(eqs, [x, y, b], expression = Val{false}) 
symbolic_solve(f, b)

ERROR: AssertionError: Invalid input

How could I do it?

1 Like
2

If you do the same thing as SymPy it seems to work fine:

using Symbolics, Groebner, Nemo, LinearAlgebra
@variables x[1:2] y[1:2] b[1:2]
eqs = collect([one.(x) x] * b .~ y)
symbolic_solve(eqs, collect(b))

1-element Vector{Any}:
 Dict{Num, Any}(b[2] => (y[1] - y[2]) / (x[1] - x[2]), b[1] => (x[1]*y[2] - x[2]*y[1]) / (x[1] - x[2]))
1 Like