Merge branch 'master' of github.com:MilesCranmer/PySR into master

TODO.md

@@ -79,6 +79,8 @@
 ## Algorithmic performance ideas:
 
 - [ ] Use package compiler and compile sr.jl into a standalone binary that can be used by pysr.
+- [ ] When doing equation warmup, only migrate those equations with almost the same complexity. Rather than having to consider simple equations later in the game.
+- [ ] Right now we only update the score based on some. Need to update score based on entire data! Note that optimizer only is used sometimes.
 - [ ] Idea: use gradient of equation with respect to each operator (perhaps simply add to each operator) to tell which part is the most "sensitive" to changes. Then, perhaps insert/delete/mutate on that part of the tree?
 - [ ] Start populations staggered; so that there is more frequent printing (and pops that start a bit later get hall of fame already)?
 - [ ] Consider adding mutation for constant<->variable
@@ -94,6 +96,7 @@
 
 ## Code performance ideas:
 
+- [ ] How hard is it to turn the recursive array evaluation into a for loop?
 - [ ] Try defining a binary tree as an array, rather than a linked list. See https://stackoverflow.com/a/6384714/2689923
 - [ ] Add true multi-node processing, with MPI, or just file sharing. Multiple populations per core.
     - Ongoing in cluster branch

docs/operators.md

@@ -8,7 +8,7 @@ of these and other valid operators are stated below.
 
 **Binary**
 
-`plus`, `mult`, `pow`, `div`, `greater`, `mod`, `beta`, `logical_or`,
+`plus`, `sub`, `mult`, `pow`, `div`, `greater`, `mod`, `beta`, `logical_or`,
 `logical_and`
 
 **Unary**

julia/operators.jl

@@ -13,6 +13,7 @@ import Base.FastMath: sqrt_llvm_fast, neg_float_fast,
 # Use some fast operators from https://github.com/JuliaLang/julia/blob/81597635c4ad1e8c2e1c5753fda4ec0e7397543f/base/fastmath.jl
 # Define allowed operators. Any julia operator can also be used.
 plus(x::Float32, y::Float32)::Float32 = add_float_fast(x, y) #Do not change the name of this operator.
+sub(x::Float32, y::Float32)::Float32 = sub_float_fast(x, y) #Do not change the name of this operator.
 mult(x::Float32, y::Float32)::Float32 = mul_float_fast(x, y) #Do not change the name of this operator.
 pow(x::Float32, y::Float32)::Float32 = sign_fast(x)*ccall(("powf",libm), Float32, (Float32,Float32), abs_fast(x), y)
 div(x::Float32, y::Float32)::Float32 = div_float_fast(x, y)

pysr/sr.py

@@ -17,6 +17,7 @@ sympy_mappings = {
     'div':  lambda x, y : x/y,
     'mult': lambda x, y : x*y,
     'plus': lambda x, y : x + y,
+    'sub':  lambda x, y : x - y,
     'neg':  lambda x    : -x,
     'pow':  lambda x, y : sympy.sign(x)*abs(x)**y,
     'cos':  lambda x    : sympy.cos(x),