Coverage for pySDC/projects/TOMS/visualize_pySDC_with_PETSc.py: 99%
93 statements
« prev ^ index » next coverage.py v7.6.7, created at 2024-11-16 14:51 +0000
« prev ^ index » next coverage.py v7.6.7, created at 2024-11-16 14:51 +0000
1import os
3import matplotlib.colors as colors
4import numpy as np
6import pySDC.helpers.plot_helper as plt_helper
9def is_number(s):
10 """
11 Helper function to detect numbers
13 Args:
14 s: a string
16 Returns:
17 bool: True if s is a number
18 """
19 try:
20 float(s)
21 return True
22 except ValueError:
23 pass
25 try:
26 import unicodedata
28 unicodedata.numeric(s)
29 return True
30 except (TypeError, ValueError):
31 pass
33 return False
36def join_timings(file=None, result=None, cwd=''):
37 """
38 Helper function to read in JUBE result tables and convert/join them into a single dictionary
40 Args:
41 file: current fils containing a JUBE result table
42 result: dictionary (empty or not)
43 cwd (str): current working directory
45 Returns:
46 dict: result dictionary for further usage
47 """
48 with open(cwd + file) as f:
49 lines = f.readlines()
51 for line in lines:
52 line_split = line.replace('\n', '').replace(' ', '').split('|')
53 if is_number(line_split[0]):
54 ntime = int(int(line_split[0]) * int(line_split[1]) / int(line_split[2]))
55 nspace = int(line_split[2])
56 timing = float(line_split[3])
57 result[(nspace, ntime)] = timing
59 return result
62def truncate_colormap(cmap, minval=0.0, maxval=1.0, n=100):
63 """
64 Helper function to crop a colormap
66 Args:
67 cmap: colormap
68 minval: minimum value
69 maxval: maximum value
70 n: stepsize
72 Returns:
73 cropped colormap
74 """
75 new_cmap = colors.LinearSegmentedColormap.from_list(
76 'trunc({n},{a:.2f},{b:.2f})'.format(n=cmap.name, a=minval, b=maxval), cmap(np.linspace(minval, maxval, n))
77 )
78 return new_cmap
81def visualize_matrix(result=None):
82 """
83 Visualizes runtimes in a matrix (cores in space vs. cores in time)
85 Args:
86 result: dictionary containing the runtimes
87 """
88 process_list = [1, 2, 4, 6, 12, 24]
89 dim = len(process_list)
90 mat = np.zeros((dim, dim))
91 tmin = 1e03
92 tmax = 0
93 for key, item in result.items():
94 mat[process_list.index(key[0]), process_list.index(key[1])] = item
95 tmin = min(tmin, item)
96 tmax = max(tmax, item)
98 plt_helper.setup_mpl()
99 plt_helper.newfig(textwidth=120, scale=1.5)
100 cmap = plt_helper.plt.get_cmap('RdYlGn_r')
101 new_cmap = truncate_colormap(cmap, 0.1, 0.9)
102 plt_helper.plt.imshow(
103 mat.T, origin='lower', norm=colors.LogNorm(vmin=tmin, vmax=tmax), cmap=new_cmap, aspect='auto'
104 )
106 for key, item in result.items():
107 timing = "{:3.1f}".format(item)
108 plt_helper.plt.annotate(
109 timing,
110 xy=(process_list.index(key[0]), process_list.index(key[1])),
111 size='x-small',
112 ha='center',
113 va='center',
114 )
116 plt_helper.plt.xticks(range(dim), process_list)
117 plt_helper.plt.yticks(range(dim), process_list)
118 plt_helper.plt.xlabel('Cores in space')
119 plt_helper.plt.ylabel('Cores in time')
121 fname = 'data/runtimes_matrix_heat'
122 plt_helper.savefig(fname)
124 assert os.path.isfile(fname + '.pdf'), 'ERROR: plotting did not create PDF file'
125 # assert os.path.isfile(fname + '.pgf'), 'ERROR: plotting did not create PGF file'
126 assert os.path.isfile(fname + '.png'), 'ERROR: plotting did not create PNG file'
129def visualize_speedup(result=None):
130 """
131 Visualizes runtimes of two different runs (MLSDC vs. PFASST)
133 Args:
134 result: dictionary containing the runtimes
135 """
136 process_list_MLSDC = [1, 2, 4, 6, 12, 24]
137 process_list_PFASST = [24, 48, 96, 144, 288, 576]
139 timing_MLSDC = np.zeros(len(process_list_MLSDC))
140 timing_PFASST = np.zeros((len(process_list_PFASST)))
141 for key, item in result.items():
142 if key[0] * key[1] in process_list_MLSDC:
143 timing_MLSDC[process_list_MLSDC.index(key[0] * key[1])] = item
144 if key[0] * key[1] in process_list_PFASST:
145 timing_PFASST[process_list_PFASST.index(key[0] * key[1])] = item
147 plt_helper.setup_mpl()
148 plt_helper.newfig(textwidth=120, scale=1.5)
150 process_list_all = process_list_MLSDC + process_list_PFASST
151 ideal = [timing_MLSDC[0] / nproc for nproc in process_list_all]
152 plt_helper.plt.loglog(process_list_all, ideal, 'k--', label='ideal')
153 plt_helper.plt.loglog(process_list_MLSDC, timing_MLSDC, 'bo-', label='MLSDC')
154 plt_helper.plt.loglog(process_list_PFASST, timing_PFASST, 'rs-', label='PFASST')
156 plt_helper.plt.xlim(process_list_all[0] / 2, process_list_all[-1] * 2)
157 plt_helper.plt.ylim(ideal[-1] / 2, ideal[0] * 2)
158 plt_helper.plt.xlabel('Number of cores')
159 plt_helper.plt.ylabel('Runtime (sec.)')
161 plt_helper.plt.legend()
162 plt_helper.plt.grid()
164 fname = 'data/speedup_heat'
165 plt_helper.savefig(fname)
166 assert os.path.isfile(fname + '.pdf'), 'ERROR: plotting did not create PDF file'
167 # assert os.path.isfile(fname + '.pgf'), 'ERROR: plotting did not create PGF file'
168 assert os.path.isfile(fname + '.png'), 'ERROR: plotting did not create PNG file'
171def main(cwd=''):
172 """
173 Main routine to call them all
175 Args:
176 cwd (str): current working directory
178 """
179 result = {}
180 files = [
181 'data/result_PFASST_1_NEW.dat',
182 'data/result_PFASST_2_NEW.dat',
183 'data/result_PFASST_4_NEW.dat',
184 'data/result_PFASST_6_NEW.dat',
185 'data/result_PFASST_12_NEW.dat',
186 'data/result_PFASST_24_NEW.dat',
187 ]
188 for file in files:
189 result = join_timings(file=file, result=result, cwd=cwd)
190 visualize_matrix(result=result)
192 result = {}
193 files = ['data/result_MLSDC_NEW.dat', 'data/result_PFASST_multinode_24_NEW.dat']
194 for file in files:
195 result = join_timings(file=file, result=result, cwd=cwd)
196 # result.pop((24, 24))
197 visualize_speedup(result=result)
200if __name__ == "__main__":
201 main()