PySR / eureqa.jl
MilesCranmer's picture
Remove example code
8d9fde2
raw
history blame
11.9 kB
# Define allowed operators
plus(x::Float64, y::Float64) = x+y
mult(x::Float64, y::Float64) = x*y;
##########################
# # Allowed operators
# (Apparently using const for globals helps speed)
const binops = [plus, mult]
const unaops = [sin, cos, exp]
##########################
const nvar = 5;
# Here is the function we want to learn (x2^2 + cos(x3) + 5)
#
##########################
# # Dataset to learn
const X = rand(100, nvar)
const y = ((cx,)->cx^2).(X[:, 2]) + cos.(X[:, 3]) .+ 5.0;
##########################
##################
# Hyperparameters
# How much to punish complexity
const parsimony = 0.01
# How much to scale temperature by (T between 0 and 1)
const alpha = 10.0
const maxsize = 20
##################
id = (x,) -> x
const nuna = size(unaops)[1]
const nbin = size(binops)[1]
const nops = nuna + nbin
# Define a serialization format for the symbolic equations:
mutable struct Node
#Holds operators, variables, constants in a tree
degree::Int #0 for constant/variable, 1 for cos/sin, 2 for +/* etc.
val::Union{Float64, Int} #Either const value, or enumerates variable
constant::Bool #false if variable
op::Function #enumerates operator (for degree=1,2)
l::Union{Node, Nothing}
r::Union{Node, Nothing}
Node(val::Float64) = new(0, val, true, id, nothing, nothing)
Node(val::Int) = new(0, val, false, id, nothing, nothing)
Node(op, l::Node) = new(1, 0.0, false, op, l, nothing)
Node(op, l::Union{Float64, Int}) = new(1, 0.0, false, op, Node(l), nothing)
Node(op, l::Node, r::Node) = new(2, 0.0, false, op, l, r)
#Allow to pass the leaf value without additional node call:
Node(op, l::Union{Float64, Int}, r::Node) = new(2, 0.0, false, op, Node(l), r)
Node(op, l::Node, r::Union{Float64, Int}) = new(2, 0.0, false, op, l, Node(r))
Node(op, l::Union{Float64, Int}, r::Union{Float64, Int}) = new(2, 0.0, false, op, Node(l), Node(r))
end
# Evaluate a symbolic equation:
function evalTree(tree::Node, x::Array{Float64, 1}=Float64[])::Float64
if tree.degree == 0
if tree.constant
return tree.val
else
return x[tree.val]
end
elseif tree.degree == 1
return tree.op(evalTree(tree.l, x))
else
return tree.op(evalTree(tree.l, x), evalTree(tree.r, x))
end
end
# Count the operators, constants, variables in an equation
function countNodes(tree::Node)::Int
if tree.degree == 0
return 1
elseif tree.degree == 1
return 1 + countNodes(tree.l)
else
return 1 + countNodes(tree.l) + countNodes(tree.r)
end
end
# Convert an equation to a string
function stringTree(tree::Node)::String
if tree.degree == 0
if tree.constant
return string(tree.val)
else
return "x$(tree.val)"
end
elseif tree.degree == 1
return "$(tree.op)($(stringTree(tree.l)))"
else
return "$(tree.op)($(stringTree(tree.l)), $(stringTree(tree.r)))"
end
end
# Print an equation
function printTree(tree::Node)
println(stringTree(tree))
end
# Return a random node from the tree
function randomNode(tree::Node)::Node
if tree.degree == 0
return tree
end
a = countNodes(tree)
b = 0
c = 0
if tree.degree >= 1
b = countNodes(tree.l)
end
if tree.degree == 2
c = countNodes(tree.r)
end
i = rand(1:1+b+c)
if i <= b
return randomNode(tree.l)
elseif i == b + 1
return tree
end
return randomNode(tree.r)
end
# Count the number of unary operators in the equation
function countUnaryOperators(tree::Node)::Int
if tree.degree == 0
return 0
elseif tree.degree == 1
return 1 + countUnaryOperators(tree.l)
else
return 0 + countUnaryOperators(tree.l) + countUnaryOperators(tree.r)
end
end
# Count the number of binary operators in the equation
function countBinaryOperators(tree::Node)::Int
if tree.degree == 0
return 0
elseif tree.degree == 1
return 0 + countBinaryOperators(tree.l)
else
return 1 + countBinaryOperators(tree.l) + countBinaryOperators(tree.r)
end
end
# Count the number of operators in the equation
function countOperators(tree::Node)::Int
return countUnaryOperators(tree) + countBinaryOperators(tree)
end
# Randomly convert an operator into another one (binary->binary;
# unary->unary)
function mutateOperator(tree::Node)::Node
if countOperators(tree) == 0
return tree
end
node = randomNode(tree)
while node.degree == 0
node = randomNode(tree)
end
if node.degree == 1
node.op = unaops[rand(1:length(unaops))]
else
node.op = binops[rand(1:length(binops))]
end
return tree
end
# Count the number of constants in an equation
function countConstants(tree::Node)::Int
if tree.degree == 0
return convert(Int, tree.constant)
elseif tree.degree == 1
return 0 + countConstants(tree.l)
else
return 0 + countConstants(tree.l) + countConstants(tree.r)
end
end
# Randomly perturb a constant
function mutateConstant(
tree::Node, T::Float64,
probNegate::Float64=0.01)::Node
# T is between 0 and 1.
if countConstants(tree) == 0
return tree
end
node = randomNode(tree)
while node.degree != 0 || node.constant == false
node = randomNode(tree)
end
bottom = 0.1
maxChange = T + 1.0 + bottom
factor = maxChange^rand()
makeConstBigger = rand() > 0.5
if makeConstBigger
node.val *= factor
else
node.val /= factor
end
if rand() > probNegate
node.val *= -1
end
return tree
end
# Evaluate an equation over an array of datapoints
function evalTreeArray(
tree::Node,
x::Array{Float64, 2})::Array{Float64, 1}
return mapslices(
(cx,) -> evalTree(tree, cx),
x,
dims=[2]
)[:, 1]
end
# Sum of square error between two arrays
function SSE(x::Array{Float64}, y::Array{Float64})::Float64
return sum(((cx,)->cx^2).(x - y))
end
# Mean of square error between two arrays
function MSE(x::Array{Float64}, y::Array{Float64})::Float64
return SSE(x, y)/size(x)[1]
end
# Score an equation
function scoreFunc(
tree::Node,
X::Array{Float64, 2},
y::Array{Float64, 1},
parsimony::Float64=0.1)::Float64
return MSE(evalTreeArray(tree, X), y) + countNodes(tree)*parsimony
end
# Add a random unary/binary operation to the end of a tree
function appendRandomOp(tree::Node)::Node
node = randomNode(tree)
while node.degree != 0
node = randomNode(tree)
end
choice = rand()
makeNewBinOp = choice < nbin/nops
if rand() > 0.5
left = randn()
else
left = rand(1:nvar)
end
if rand() > 0.5
right = randn()
else
right = rand(1:nvar)
end
if makeNewBinOp
newnode = Node(
binops[rand(1:length(binops))],
left,
right
)
else
newnode = Node(
unaops[rand(1:length(unaops))],
left
)
end
node.l = newnode.l
node.r = newnode.r
node.op = newnode.op
node.degree = newnode.degree
node.val = newnode.val
node.constant = newnode.constant
return tree
end
# Select a random node, and replace it an the subtree
# with a variable or constant
function deleteRandomOp(tree::Node)::Node
node = randomNode(tree)
# Can "delete" variable or constant too
if rand() > 0.5
val = randn()
else
val = rand(1:nvar)
end
newnode = Node(val)
node.l = newnode.l
node.r = newnode.r
node.op = newnode.op
node.degree = newnode.degree
node.val = newnode.val
node.constant = newnode.constant
return tree
end
# Go through one simulated annealing mutation cycle
# exp(-delta/T) defines probability of accepting a change
function iterate(
tree::Node, T::Float64,
X::Array{Float64, 2}, y::Array{Float64, 1},
alpha::Float64=1.0,
mult::Float64=0.1
)::Node
prev = deepcopy(tree)
mutationChoice = rand()
weights = [8, 1, 1, 1]
weights /= sum(weights)
cweights = cumsum(weights)
n = countNodes(tree)
if mutationChoice < cweights[1]
tree = mutateConstant(tree, T)
elseif mutationChoice < cweights[2]
tree = mutateOperator(tree)
elseif mutationChoice < cweights[3] && n < maxsize
tree = appendRandomOp(tree)
elseif mutationChoice < cweights[4]
tree = deleteRandomOp(tree)
end
try
beforeLoss = scoreFunc(prev, X, y, mult)
afterLoss = scoreFunc(tree, X, y, mult)
delta = afterLoss - beforeLoss
probChange = exp(-delta/(T*alpha))
if probChange > rand()
return tree
end
return prev
catch error
# Sometimes too many chained exp operators
if isa(error, DomainError)
return prev
else
throw(error)
end
end
end
# Create a random equation by appending random operators
function genRandomTree(length::Int)::Node
tree = Node(1.0)
for i=1:length
tree = appendRandomOp(tree)
end
return tree
end
# Define a member of population by equation, score, and age
mutable struct PopMember
tree::Node
score::Float64
birth::Float64
PopMember(t) = new(t, scoreFunc(t, X, y, parsimony), time()-1.6e9)
end
# A list of members of the population, with easy constructors,
# which allow for random generation of new populations
mutable struct Population
members::Array{PopMember, 1}
n::Int
Population(pop::Array{PopMember, 1}) = new(pop, size(pop)[1])
Population(npop::Int64) = new([PopMember(genRandomTree(3)) for i=1:npop], npop)
Population(npop::Int64, nlength::Int64) = new([PopMember(genRandomTree(nlength)) for i=1:npop], npop)
end
# Sample 10 random members of the population, and make a new one
function samplePop(pop::Population)::Population
idx = rand(1:pop.n, 10)
return Population(pop.members[idx])#Population(deepcopy(pop.members[idx]))
end
# Sample the population, and get the best member from that sample
function bestOfSample(pop::Population)::PopMember
sample = samplePop(pop)
best_idx = argmin([sample.members[member].score for member=1:sample.n])
return sample.members[best_idx]
end
# Return best 10 examples
function bestSubPop(pop::Population)::Population
best_idx = sortperm([pop.members[member].score for member=1:pop.n])
return Population(pop.members[best_idx[1:10]])
end
# Mutate the best sampled member of the population
function iterateSample(pop::Population, T::Float64)::PopMember
allstar = bestOfSample(pop)
new = iterate(allstar.tree, T, X, y, alpha, parsimony)
allstar.tree = new
allstar.score = scoreFunc(new, X, y, parsimony)
allstar.birth = time() - 1.6e9
return allstar
end
# Pass through the population several times, replacing the oldest
# with the fittest of a small subsample
function regEvolCycle(pop::Population, T::Float64)::Population
for i=1:Int(pop.n/10)
baby = iterateSample(pop, T)
#printTree(baby.tree)
oldest = argmin([pop.members[member].birth for member=1:pop.n])
pop.members[oldest] = baby
end
return pop
end
# Cycle through regularized evolution many times,
# printing the fittest equation every 10% through
function run(
pop::Population,
ncycles::Int,
annealing::Bool=false,
)::Population
pop = deepcopy(pop)
allT = LinRange(1.0, 0.0, ncycles)
for iT in 1:size(allT)[1]
if annealing
pop = regEvolCycle(pop, allT[iT])
else
pop = regEvolCycle(pop, 1.0)
end
end
return pop
end