Missing #include<atomic>

Added python part of SAR experiment.

Author: highperformancecoder

models/SAR.py

@@ -0,0 +1,81 @@
+# run a single instance of a SAR ensemble run
+from ecolab_model import spatial_ecolab as ecolab
+
+from random import random, seed as randomSeed
+
+from ecolab import array_urand, myid, device
+
+import os
+import sys
+from math import sqrt
+
+# we want initialisation to be identical across all processes
+randomSeed(1)
+
+# we want the array operations to have a different seed across processes
+array_urand.seed(10+myid())
+
+
+nsp=int(sys.argv[1])         # initial number of species
+A=int(sys.argv[2])           # area
+mut_max=float(sys.argv[3])   # mutation rate
+migration=float(sys.argv[4]) # initial migration rate
+
+
+ecolab.repro_min(-0.1)
+ecolab.repro_max(0.1)
+ecolab.odiag_min(-1e-5)
+ecolab.odiag_max(1e-5)
+ecolab.mut_max(mut_max)
+ecolab.sp_sep(0.1)
+
+def randomList(num, min, max):
+    return [random()*(max-min)+min for i in range(num)]
+
+ecolab.species(range(nsp))
+
+numX=int(sqrt(A))
+numY=A//numX
+ecolab.setGrid(numX,numY)
+ecolab.partitionObjects()
+
+# initialises allocators and space on GPU, so that density arrays can be set
+#ecolab.makeConsistent()
+
+for i in range(numX):
+    for j in range(numY):
+        ecolab.cell(i,j).density(nsp*[100])
+        
+ecolab.repro_rate(randomList(nsp, ecolab.repro_min(), ecolab.repro_max()))
+ecolab.interaction.diag(randomList(nsp, -1e-3, -1e-3))
+ecolab.random_interaction(3,0)
+              
+ecolab.interaction.val(randomList(len(ecolab.interaction.val), ecolab.odiag_min(), ecolab.odiag_max()))
+
+ecolab.mutation(nsp*[mut_max])
+ecolab.migration(nsp*[migration])
+                  
+from plot import plot
+from GUI import gui, statusBar, windows
+
+def stepImpl():
+    ecolab.generate(100)
+    ecolab.mutate()
+    ecolab.migrate()
+    ecolab.condense()
+    ecolab.gather()
+    
+def av(x):
+    return sum(x)/len(x) if len(x)>0 else 0
+    
+def step():
+    stepImpl()
+    # area, mutation, migration, no. species, connectivity
+
+for i in range(100):
+    step()
+
+print(numX*numY, av(ecolab.mutation()), av(ecolab.migration()), len(ecolab.species), sum([x*x for x in ecolab.interaction.val()])/len(ecolab.species)**2)
+
+
+

models/starComplexity.cc

@@ -8,6 +8,7 @@
 #include "ecolab_epilogue.h"
 
 #include <algorithm>
+#include <atomic>
 #include <assert.h>
 #include <time.h>