Merge pull request #335 from Neuroblox/add-simpler-RF-learning-example

add a simple RF example

examples/RF_learning_simple.jl

@@ -0,0 +1,77 @@
+using CSV
+using DataFrames
+using Neuroblox
+using DifferentialEquations
+using MetaGraphs
+
+
+time_block_dur = 90 # ms (size of discrete time blocks)
+N_trials = 10 #number of trials
+
+global_ns = :g 
+
+fn = joinpath(@__DIR__, "image_example.csv") #stimulus image file
+data = CSV.read(fn, DataFrame)
+
+#define the circuit blox
+@named stim = ImageStimulus(data[1:N_trials,:]; namespace=global_ns, t_stimulus=600, t_pause=1000) 
+
+@named LC = NextGenerationEIBlox(;namespace=global_ns, Cₑ=2*26,Cᵢ=1*26, Δₑ=0.5, Δᵢ=0.5, η_0ₑ=10.0, η_0ᵢ=0.0, v_synₑₑ=10.0, v_synₑᵢ=-10.0, v_synᵢₑ=10.0, v_synᵢᵢ=-10.0, alpha_invₑₑ=10.0/26, alpha_invₑᵢ=0.8/26, alpha_invᵢₑ=10.0/26, alpha_invᵢᵢ=0.8/26, kₑₑ=0.0*26, kₑᵢ=0.6*26, kᵢₑ=0.6*26, kᵢᵢ=0*26) 
+
+@named VC = CorticalBlox(N_wta=45, N_exci=5,  density=0.01, weight=1,I_bg_ar=0;namespace=global_ns) 
+
+@named PFC = CorticalBlox(N_wta=20, N_exci=5, density=0.01, weight=1,I_bg_ar=0;namespace=global_ns) 
+
+@named STR1 = Striatum(N_inhib=1;namespace=global_ns) 
+@named STR2 = Striatum(N_inhib=1;namespace=global_ns) 
+
+@named tan_pop1 = TAN(;namespace=global_ns) 
+@named tan_pop2 = TAN(;namespace=global_ns) 
+
+@named AS = GreedyPolicy(namespace=global_ns, t_decision=180.5) 
+@named SNcb = SNc(namespace=global_ns) 
+
+
+assembly = [LC, VC, PFC, STR1, STR2, stim, tan_pop1, tan_pop2, AS, SNcb]
+d = Dict(b => i for (i,b) in enumerate(assembly))
+
+#define learning rules
+hebbian_mod = HebbianModulationPlasticity(K=0.04, decay=0.01, α=2.5, θₘ=1, modulator=SNcb, t_pre=1600-eps(), t_post=1600-eps(), t_mod=90)
+
+hebbian_cort = HebbianPlasticity(K=5e-4, W_lim=5, t_pre=1600-eps(), t_post=1600-eps())
+
+hebbian_thal_cort = HebbianPlasticity(K=1.7e-5, W_lim=6, t_pre=1600-eps(), t_post=1600-eps())
+
+g = MetaDiGraph()
+add_blox!.(Ref(g), assembly)
+
+#connect the bloxs
+
+add_edge!(g, d[LC], d[VC], Dict(:weight => 44)) #LC->VC
+add_edge!(g, d[LC], d[PFC], Dict(:weight => 44)) #LC->pfc
+
+add_edge!(g, d[VC], d[PFC], Dict(:weight => 1, :density => 0.08, :learning_rule => hebbian_cort)) #VC->pfc
+add_edge!(g, d[PFC], d[STR1], Dict(:weight => 0.075, :density => 0.04, :learning_rule => hebbian_mod)) #pfc->str1
+add_edge!(g, d[PFC], d[STR2], Dict(:weight => 0.075, :density => 0.04, :learning_rule => hebbian_mod)) #pfc->str2
+
+add_edge!(g, d[stim], d[VC], :weight, 14) #stim->VC
+add_edge!(g, d[tan_pop1], d[STR1], Dict(:weight => 1, :t_event => 90.0)) #TAN pop1 -> str1
+add_edge!(g, d[tan_pop2], d[STR2], Dict(:weight => 1, :t_event => 90.0)) #TAN pop2 -> str2
+add_edge!(g, d[STR1], d[tan_pop1], Dict(:weight => 1)) #str1 -> TAN pop1 
+add_edge!(g, d[STR2], d[tan_pop1], Dict(:weight => 1)) #str2 -> TAN pop1
+add_edge!(g, d[STR1], d[tan_pop2], Dict(:weight => 1)) #str1 -> TAN pop2 
+add_edge!(g, d[STR2], d[tan_pop2], Dict(:weight => 1)) #str2 -> TAN pop2
+add_edge!(g, d[STR1], d[STR2], Dict(:weight => 1, :t_event => 181.0)) #str1 -> str2
+add_edge!(g, d[STR2], d[STR1], Dict(:weight => 1, :t_event => 181.0)) #str2 -> str1
+add_edge!(g, d[STR1], d[AS])# str1->AS
+add_edge!(g, d[STR2], d[AS])# str2->AS
+add_edge!(g, d[STR1], d[SNcb], Dict(:weight => 1)) # str1->Snc
+add_edge!(g, d[STR2], d[SNcb], Dict(:weight => 1))  # str2->Snc
+
+#define the circuit as an Agent
+agent = Agent(g; name=:ag, t_block = 90);
+
+#define environment : contains stimuli and feedback
+env = ClassificationEnvironment(stim; name=:env, namespace=global_ns)
+
+run_experiment!(agent, env; alg=Vern7(), reltol=1e-9,abstol=1e-9)