#Create a simplified model.  Must select the appropriate list
#containing the simplified set of predictor variables - see
#previous 'dismo' steps.

myGbmModelOut.abi.bal.simplified <- gbm.step(
data=myData,
gbm.x = myExpl,
gbm.y = myRespName.abi.bal,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.005,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.ace.rub.simplified <- gbm.step(
data=myData,
gbm.x = myExpl,
gbm.y = myRespName.ace.rub,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.003,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.ace.sac.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.ace.sac$pred.list[[4]],
gbm.y = myRespName.ace.sac,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.025,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.aln.inc.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.aln.inc$pred.list[[2]],
gbm.y = myRespName.aln.inc,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.0075,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.bet.all.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.bet.all$pred.list[[2]],
gbm.y = myRespName.bet.all,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.0075,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.car.spp.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.car.spp$pred.list[[1]],
gbm.y = myRespName.car.spp,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.01,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.cas.den.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.cas.den$pred.list[[2]],
gbm.y = myRespName.cas.den,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.01,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.fag.gra.simplified <- gbm.step(
data=myData,
gbm.x = myExpl,
gbm.y = myRespName.fag.gra,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.015,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.fra.ame.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.fra.ame$pred.list[[1]],
gbm.y = myRespName.fra.ame,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.02,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.fra.nig.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.fra.nig$pred.list[[1]],
gbm.y = myRespName.fra.nig,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.005,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.jug.cin.simplified <- gbm.step(
data=myData,
gbm.x = myExpl,
gbm.y = myRespName.jug.cin,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.01,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.jug.nig.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.jug.nig$pred.list[[1]],
gbm.y = myRespName.jug.nig,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.0015,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.lir.tul.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.lir.tul$pred.list[[2]],
gbm.y = myRespName.lir.tul,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.0075,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.mag.acu.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.mag.acu$pred.list[[5]],
gbm.y = myRespName.mag.acu,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.015,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.ost.vir.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.ost.vir$pred.list[[2]],
gbm.y = myRespName.ost.vir,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.0025,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.pin.str.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.pin.str$pred.list[[1]],
gbm.y = myRespName.pin.str,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.02,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.pla.occ.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.pla.occ$pred.list[[2]],
gbm.y = myRespName.pla.occ,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.0025,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.pru.ser.simplified <- gbm.step(
data=myData,
gbm.x = myExpl,
gbm.y = myRespName.pru.ser,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.005,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.que.alb.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.que.alb$pred.list[[1]],
gbm.y = myRespName.que.alb,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.0075,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.que.vel.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.que.vel$pred.list[[2]],
gbm.y = myRespName.que.vel,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.02,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.til.ame.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.til.ame$pred.list[[8]],
gbm.y = myRespName.til.ame,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.02,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.tsu.can.simplified <- gbm.step(
data=myData,
gbm.x = myExpl,
gbm.y = myRespName.tsu.can,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.02,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)

myGbmModelOut.ulm.ame.simplified <- gbm.step(
data=myData,
gbm.x = myGbmModelOut.simplify.variables.ulm.ame$pred.list[[1]],
gbm.y = myRespName.ulm.ame,
family = "bernoulli",
tree.complexity = 5,
learning.rate = 0.01,
n.trees = 500,
max.trees = 2500,
step.size = 50,
bag.fraction = 0.5)



#Save workspace.

save.image(file="sdm.objects.with.Native.variables.biomod2.step.3.RData")