Flux.jl/src/compiler/loops.jl
2017-03-14 15:21:18 +00:00

135 lines
3.3 KiB
Julia

export unroll, unroll1
type Offset
name::Symbol
n::Int
default::Nullable{Param}
end
Offset(name, n) = Offset(name, n, nothing)
Base.:-(o::Offset) = Offset(o.name, -o.n, o.default)
function liftloops(ex)
ex = DataFlow.normedges(ex)
decls = Dict()
ex = MacroTools.postwalk(ex) do ex
@capture(ex, x_{n_}) || return ex
haskey(decls, (x,n)) && return namify(decls[(x,n)])
@gensym edge
decls[(x,n)] = :($edge = $(Offset(x,n))($x))
edge
end
prepend!(ex.args, collect(values(decls)))
ex
end
function hasloops(model)
g = graph(model)
g == nothing && return false
iscyclic(g) && return true
result = false
map(m -> hasloops(m) && (result = true), g)
return result
end
function atomise(model)
postwalk(graph(model)) do v
hasloops(value(v)) || return v
spliceinputs(atomise(value(v)), inputs(v)...)
end
end
function collect_state(v::IVertex)
state = typeof(v)[]
offset = Int[]
default = Param[]
prewalk!(v) do v
value(v) isa Offset || return v
if (i = findfirst(state, v[1])) == 0
push!(state, v[1])
push!(offset, max(0, -value(v).n))
push!(default, get(value(v).default))
else
offset[i] = max(offset[i], -value(v).n)
end
v
end
return state, offset, default
end
hiddeninput(n) = vertex(Split(n), inputnode(1))
function create_steps(v::IVertex, n; seq = true, stateful = true)
[(stateful ? bumpinputs : copy)(seq ? spliceinputs(v, hiddeninput(i)) : v) for i = 1:n]
end
function getvar(n, step, steps, offset, default; stateful = true)
if stateful && step < 1
hiddeninput(sum(offset[1:n-1]) + 1 - step)
elseif step 1:length(steps)
constant(default[n])
else
steps[step][1,n]
end
end
function stateout(steps, offset, default)
outs = []
defaults = []
for i = 1:length(offset), j = 1:offset[i]
push!(outs, getvar(i, length(steps)-j+1, steps, offset, default))
push!(defaults, default[i])
end
group(outs...), defaults
end
function unrollgraph(v::IVertex, n; seq = true, stateful = true)
state, offset, default = collect_state(v)
v = group(group(state...), v)
steps = create_steps(v, n, seq = seq, stateful = stateful)
for i = 1:n
vars = inputs(steps[i][1])
postwalk!(steps[i]) do v
value(v) isa Offset || return v
varid = findfirst(vars,v[1])
getvar(varid, value(v).n + i, steps, offset, default, stateful = stateful)
end
end
out = group(map(x->x[2], steps)...)
if stateful
state, defaults = stateout(steps, offset, default)
group(state,out), map(Flux.state, defaults)
else
out, []
end
end
unrollgraph(m, n; kws...) = unrollgraph(atomise(m), n; kws...)
# TODO: perhaps split into SeqModel + StatefulModel
type Unrolled <: Model
model
graph::IVertex{Any}
state::Vector{Any}
stateful::Bool
steps::Int
end
(m::Unrolled)(xs...) = interpret(reifyparams(m.graph), xs...)
graph(u::Unrolled) = u.graph
function unroll(model, n; seq = true, stateful = true)
graph, state = unrollgraph(model, n; seq = seq, stateful = stateful)
seq || stateful ? Unrolled(model, graph, state, stateful, n) : Capacitor(graph)
end
function unroll1(model)
graph, state = unrollgraph(model, 1; seq = false)
graph = group(graph[1], map(x->x[1], inputs(graph)[2:end])...)
Unrolled(model, graph, state, false, 1)
end
flip(model) = Capacitor(map(x -> x isa Offset ? -x : x, atomise(model)))