8f03078 · opencog-CompactHash

Changed files (6) +177 -0

do_plots/do_plot.py +71 -0

…	…	@@ -0,0 +1,71 @@
	1	from matplotlib import pyplot
	2	import string
	3	from array import *
	4
	5	fp = open("memopt.dat",'r')
	6
	7	X1 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	8	Y1 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	9	Y2 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	10	Y3 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	11	Y4 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	12	Y5 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	13	Y6 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	14	Y7 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	15	Y8 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	16	Y9 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	17	Y10 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	18	Y11 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	19
	20	line = fp.readline()
	21
	22	for i in range(0,7):
	23	line = fp.readline()
	24	value = line.split(",")
	25	X1[i] = float(value[0])
	26	Y1[i] = float(value[1])
	27	Y2[i] = float(value[2])
	28	Y3[i] = float(value[3])
	29	Y4[i] = float(value[4])
	30	Y5[i] = float(value[5])
	31	Y6[i] = float(value[6])
	32	Y7[i] = float(value[7])
	33	Y8[i] = float(value[8])
	34	Y9[i] = float(value[9])
	35	Y10[i] = float(value[10])
	36	Y11[i] = float(value[11])
	37	Y3[i] /= Y1[i]
	38	Y4[i] /= Y1[i]
	39	Y5[i] /= Y1[i]
	40	Y6[i] /= Y1[i]
	41	Y7[i] /= Y1[i]
	42	Y8[i] /= Y1[i]
	43	Y9[i] /= Y1[i]
	44	Y10[i] /= Y1[i]
	45	Y11[i] /= Y1[i]
	46	#print (X1[i]pow(2,i+1) X1[i]*pow(2,i+1))/Y1[i]
	47
	48	fig = pyplot.figure()
	49
	50	fig.patch.set_facecolor('white')
	51	pyplot.xlabel("Compressibility",fontdict={'fontsize':16})
	52	pyplot.ylabel("Runtime (millisecs)/numcells",fontdict={'fontsize':16})
	53
	54	pyplot.semilogy()
	55
	56	pyplot.plot(Y2, Y3, color = '#3399FF', linestyle='-', linewidth=1.0, marker='o', label='kDtree')
	57	pyplot.plot(Y2, Y4, color = 'green', linestyle='-', linewidth=1.0, marker='o', label='Full Perfect on CPU')
	58	pyplot.plot(Y2, Y5, color = 'orange', linestyle='-', linewidth=1.0, marker='o', label='7 write, 1 read on CPU')
	59	pyplot.plot(Y2, Y6, color = 'blue', linestyle='-', linewidth=1.0, marker='o', label='1 write, 3 read on CPU')
	60	pyplot.plot(Y2, Y7, color = 'red', linestyle='-', linewidth=1.0, marker='o', label='Compact CPU')
	61	pyplot.plot(Y2, Y8, color = 'green', linestyle='--', linewidth=1.0, marker='D', label='Full Perfect on GPU')
	62	pyplot.plot(Y2, Y9, color = 'orange', linestyle='--', linewidth=1.0, marker='D', label='7 write, 1 read on GPU')
	63	pyplot.plot(Y2, Y10, color = 'blue', linestyle='--', linewidth=1.0, marker='D', label='1 write, 3 read on GPU')
	64	pyplot.plot(Y2, Y11, color = 'red', linestyle='--', linewidth=1.0, marker='D', label='Compact GPU')
	65
	66	pyplot.grid()
	67	pyplot.legend(loc = 2)
	68	pyplot.suptitle("Comparison of Neighbor Finding Algorithms on the CPU and GPU",fontdict={'fontsize':16})
	69
	70	pyplot.savefig('neighbors_all.pdf',format='pdf')
	71	pyplot.show()

do_plots/do_plot_cpu.py +55 -0

…	…	@@ -0,0 +1,55 @@
	1	from matplotlib import pyplot
	2	import string
	3	from array import *
	4
	5	fp = open("memopt.dat",'r')
	6
	7	X1 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	8	Y1 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	9	Y2 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	10	Y3 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	11	Y4 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	12	Y5 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	13	Y6 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	14	Y7 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	15
	16	line = fp.readline()
	17
	18	for i in range(0,7):
	19	line = fp.readline()
	20	value = line.split(",")
	21	X1[i] = float(value[0])
	22	Y1[i] = float(value[1])
	23	Y2[i] = float(value[2])
	24	Y3[i] = float(value[3])
	25	Y4[i] = float(value[4])
	26	Y5[i] = float(value[5])
	27	Y6[i] = float(value[6])
	28	Y7[i] = float(value[7])
	29	Y3[i] /= Y1[i]
	30	Y4[i] /= Y1[i]
	31	Y5[i] /= Y1[i]
	32	Y6[i] /= Y1[i]
	33	Y7[i] /= Y1[i]
	34	#print (X1[i]pow(2,i+1) X1[i]*pow(2,i+1))/Y1[i]
	35
	36	fig = pyplot.figure()
	37
	38	fig.patch.set_facecolor('white')
	39	pyplot.xlabel("Compressibility",fontdict={'fontsize':16})
	40	pyplot.ylabel("Runtime (millisecs)/numcells",fontdict={'fontsize':16})
	41
	42	pyplot.semilogy()
	43
	44	pyplot.plot(Y2, Y3, color = '#3399FF', linestyle='-', linewidth=1.0, marker='o', label='kDtree')
	45	pyplot.plot(Y2, Y4, color = 'green', linestyle='-', linewidth=1.0, marker='o', label='Full Perfect on CPU')
	46	pyplot.plot(Y2, Y5, color = 'orange', linestyle='-', linewidth=1.0, marker='o', label='7 write, 1 read on CPU')
	47	pyplot.plot(Y2, Y6, color = 'blue', linestyle='-', linewidth=1.0, marker='o', label='1 write, 3 read on CPU')
	48	pyplot.plot(Y2, Y7, color = 'red', linestyle='-', linewidth=1.0, marker='o', label='Compact CPU')
	49
	50	pyplot.grid()
	51	pyplot.legend(loc = 2)
	52	pyplot.suptitle("Comparison of Neighbor Finding Algorithms on the CPU",fontdict={'fontsize':16})
	53
	54	pyplot.savefig('neighbors_cpu.pdf',format='pdf')
	55	pyplot.show()

do_plots/do_plot_gpu.py +51 -0

…	…	@@ -0,0 +1,51 @@
	1	from matplotlib import pyplot
	2	import string
	3	from array import *
	4
	5	fp = open("memopt.dat",'r')
	6
	7	X1 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	8	Y1 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	9	Y2 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	10	Y8 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	11	Y9 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	12	Y10 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	13	Y11 = array ('d', [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ])
	14
	15	line = fp.readline()
	16
	17	for i in range(0,7):
	18	line = fp.readline()
	19	value = line.split(",")
	20	X1[i] = float(value[0])
	21	Y1[i] = float(value[1])
	22	Y2[i] = float(value[2])
	23	Y8[i] = float(value[8])
	24	Y9[i] = float(value[9])
	25	Y10[i] = float(value[10])
	26	Y11[i] = float(value[11])
	27	Y8[i] /= Y1[i]
	28	Y9[i] /= Y1[i]
	29	Y10[i] /= Y1[i]
	30	Y11[i] /= Y1[i]
	31	#print (X1[i]pow(2,i+1) X1[i]*pow(2,i+1))/Y1[i]
	32
	33	fig = pyplot.figure()
	34
	35	fig.patch.set_facecolor('white')
	36	pyplot.xlabel("Compressibility",fontdict={'fontsize':16})
	37	pyplot.ylabel("Runtime (millisecs)/numcells",fontdict={'fontsize':16})
	38
	39	pyplot.semilogy()
	40
	41	pyplot.plot(Y2, Y8, color = 'green', linestyle='-', linewidth=1.0, marker='D', label='Full Perfect on GPU')
	42	pyplot.plot(Y2, Y9, color = 'orange', linestyle='-', linewidth=1.0, marker='D', label='7 write, 1 read on GPU')
	43	pyplot.plot(Y2, Y10, color = 'blue', linestyle='-', linewidth=1.0, marker='D', label='1 write, 3 read on GPU')
	44	pyplot.plot(Y2, Y11, color = 'red', linestyle='-', linewidth=1.0, marker='D', label='Compact GPU')
	45
	46	pyplot.grid()
	47	pyplot.legend(loc = 2)
	48	pyplot.suptitle("Comparison of Neighbor Finding Algorithms on the GPU",fontdict={'fontsize':16})
	49
	50	pyplot.savefig('neighbors_gpu.pdf',format='pdf')
	51	pyplot.show()

do_plots/neighbors_all.pdf +0 -0

(no content changes — binary or rename)

do_plots/neighbors_cpu.pdf +0 -0

(no content changes — binary or rename)

do_plots/neighbors_gpu.pdf +0 -0

(no content changes — binary or rename)

opencog-CompactHash commits / 8f03078

Changed files (6) +177 -0