9 from mvpa.misc.plot.base import plot_bars
11 # uniform colors for OS results
12 os_colors = ['#B63537', '#4E4DA0', '#008200', 'gray']
13 os_order = ['linux', 'mac', 'win', 'otheros']
15 resource_categories = {
16 'vendor': 'Vendor/Project website',
17 'retailer': 'Retailer',
18 'os': 'Operating system',
23 'freebsdports': 'FreeBSD ports',
25 'macports': 'Macports',
26 'matlabcentral': 'Matlab Central',
27 'neurodebian': 'NeuroDebian',
30 'pythonbundles': 'Python bundles',
31 'sourceforge': 'Sourceforge',
32 'other': 'Other resource'
36 'general': 'General computing',
37 'dc': 'Distributed computing',
38 'img': 'Brain imaging',
39 'datamanage': 'Data management',
40 'neusys': 'Neural systems modeling',
41 'electro': 'Electrophysiology, MEG/EEG',
42 'bci': 'Brain-computer interface',
43 'acq': 'Hardware interface/Data acquisition',
44 'rt': 'Real-time solutions',
45 'psychophys': 'Psychophysics/Experiment control'
48 # some meaningful groups of OSes
49 redhat_family = ["rhel", "centos", "fedora", "scilinux"]
50 debian_family = ["debian", "ubuntu", "biolinux"]
51 suse_family = ["suse", "slel"]
52 other_linux_family = ["gentoo", "mandriva", "arch", "slackware", "otherlinux"]
53 other_family = ["starbsd", "unix", "qnx", "beos", "solaris", "other"]
65 'linux': redhat_family + debian_family + suse_family + other_linux_family,
66 'otheros': other_family
68 # end the reverse mapping
71 for os in os_family[ost]:
72 os_family_rev[os] = ost
75 def load_list2dict(name):
80 d[kv[0]] = kv[1].strip().strip('"')
86 os_dict = load_list2dict('oslist.txt')
87 datamod_dict = load_list2dict('datamodlist.txt')
88 sw_dict = load_list2dict('swlist.txt')
90 def __init__(self, srcdir):
91 # eats the whole directory
94 datafilenames = glob('%s/*.json' % srcdir)
95 for dfn in datafilenames:
96 rawdata = json.load(open(dfn))
97 self[rawdata['timestamp']] = rawdata
99 def get_unique(self, key):
100 # return a set of all (unique) values for a field id
102 for d in self.values():
105 if isinstance(el, list):
106 uniq = uniq.union(el)
108 uniq = uniq.union((el,))
111 def get_not_none(self, key):
112 # return a list of all values of a specific field id
113 # the second return value is count of submission that did not have data
117 for d in self.values():
120 if isinstance(el, list):
131 def get_counts(self, key):
132 # return a dict with field values as keys and respective submission
134 vals = self.get_not_none(key)[0]
135 uniq = np.unique(vals)
136 counts = dict(zip(uniq, [vals.count(u) for u in uniq]))
139 def select_match(self, key, values):
140 # return a db with all submissions were a field id has one of the
143 for k, v in self.items():
147 if isinstance(el, list):
148 if len(set(values).intersection(el)):
154 def get_nice_name(self, id):
155 srcs = [DB.os_dict, os_cat_names, DB.sw_dict, sw_categories,
160 # not found, nothing nicer
164 def mkpic_os_per_env(db, destdir):
165 envs = ['pers_os', 'man_os', 'virt_host_os', 'virt_guest_os']
166 env_names = ['Personal', 'Managed', 'Virt. Host', 'Virt. Guest']
170 counts = db.get_counts(env)
171 stats = dict(zip(os_family.keys(), [0] * len(os_family)))
173 stats[os_family_rev[os]] += counts[os]
174 total_count = np.sum(stats.values())
176 stats[osf] = float(stats[osf]) / total_count
177 env_stats[env] = stats
178 # make stacked barplot
179 pl.figure(figsize=(6.4, 4.8), facecolor='w', edgecolor='k')
180 x = np.arange(len(envs))
181 bottoms = np.zeros(len(envs))
182 for i, os in enumerate(os_order):
183 stat = [env_stats[e][os] for e in envs]
184 pl.bar(x, stat, bottom=bottoms, color=os_colors[i],
185 label=db.get_nice_name(os), width=0.8)
187 pl.legend(loc='lower right')
188 pl.xticks(x + 0.4, [db.get_nice_name(e) for e in env_names])
189 pl.xlim(-0.25, len(envs))
190 pl.title("Operating system preference by environment")
191 pl.ylabel("Fraction of submissions")
192 pl.savefig('%s/ospref_by_env.png' % destdir, format='png', dpi=80)
194 def mkpic_submissions_per_datamod(db, destdir):
196 spd = db.get_counts('bg_datamod')
197 spd = sorted(spd.items(), cmp=lambda x, y: cmp(x[1], y[1]))[::-1]
198 x = np.arange(len(spd))
199 pl.figure(figsize=(6.4, 4.8), facecolor='w', edgecolor='k')
200 pl.title('Submissions per data modality')
201 pl.bar(x, [s[1] for s in spd])
202 pl.xticks(x + 0.5, [db.datamod_dict[k[0]] for k in spd], rotation=-10)
203 pl.ylabel('Survey submissions per data modality\n(multiple choices per submission possible)')
204 pl.savefig('%s/submissions_per_datamod.png' % destdir, format='png', dpi=80)
206 def mkpic_resources(db, destdir):
207 res = db.get_counts('software_resource')
209 x = np.arange(len(res))
210 pl.figure(figsize=(6.4, 4.8), facecolor='w', edgecolor='k')
211 pl.title('Software resources')
212 pl.bar(x + (1./8), [s[1] for s in res], width=0.75, color = '#008200')
213 pl.xticks(x + 0.5, ['' for s in res])
214 for i, s in enumerate(res):
215 pl.text(i+.5, 0.1, db.get_nice_name(s[0]), rotation=90,
216 horizontalalignment='center',
217 verticalalignment='bottom',
218 bbox=dict(facecolor='white', alpha=0.8, edgecolor='white',
220 pl.ylabel('Number of submissions')
221 pl.savefig('%s/software_resources.png' % destdir, format='png', dpi=80)
223 def mkpic_software(db, destdir):
224 for typ in sw_categories.keys():
225 counts = db.get_counts('sw_%s' % typ)
226 pl.figure(figsize=(6.4, 4.8), facecolor='w', edgecolor='k')
227 pl.title("Software popularity: %s" % db.get_nice_name(typ))
229 pl.text(.5, .5, "[Insufficient data for this figure]",
230 horizontalalignment='center')
234 stats = sorted(counts.items(), cmp=lambda x, y: cmp(x[0], y[0]))
235 x = np.arange(len(stats))
236 pl.bar(x + (1./8), [s[1] for s in stats], width=0.75, color = '#008200')
237 pl.xticks(x + 0.5, ['' for s in stats])
238 for i, s in enumerate(stats):
239 pl.text(i+.5, 0.1, db.get_nice_name(s[0]), rotation=90,
240 horizontalalignment='center',
241 verticalalignment='bottom',
242 bbox=dict(facecolor='white', alpha=0.8, edgecolor='white',
244 pl.xlim(0, len(stats))
245 pl.ylabel("Number of submissions")
246 pl.savefig('%s/sw_%s.png' % (destdir, typ), format='png', dpi=80)
248 def mkpic_rating_by_os(db, env, items, destdir, title):
249 pl.figure(figsize=(6.4, 4.8), facecolor='w', edgecolor='k')
250 for i, os in enumerate(os_order):
251 ratings = [db.select_match(env,
252 os_family[os]).get_not_none('%s' % (it,))[0]
254 plot_bars(ratings, offset=((i+1)*0.2)-0.1, color=os_colors[i],
255 title=title, ylabel="Mean rating", label=db.get_nice_name(os))
257 pl.xlim((0,len(items)))
258 pl.yticks((0, 3), ['Disagree', 'Agree'], rotation=90)
259 pl.xticks(np.arange(len(items))+0.5, [i[-2:] for i in items],
260 horizontalalignment='center')
261 pl.legend(loc='lower right')
262 pl.savefig('%s/ratings_%s.png' % (destdir, env), format='png', dpi=80)
265 def main(srcdir, destdir):
267 for pic in [mkpic_submissions_per_datamod, mkpic_os_per_env, mkpic_software,
270 mkpic_rating_by_os(db, 'pers_os', ['pers_r%i' % i for i in range(1, 9)], destdir,
271 "Ratings: Personal environment")
272 mkpic_rating_by_os(db, 'man_os', ['man_r%i' % i for i in range(1, 5)], destdir,
273 "Ratings: Managed environment")
274 mkpic_rating_by_os(db, 'virt_host_os', ['virt_r%i' % i for i in range(1, 4)], destdir,
275 "Ratings: Virtual environment (by host OS)")
276 # submission stats: this is RST
277 statsfile = open('%s/stats.txt' % destdir, 'w')
278 statsfile.write('::\n\n Number of submissions: %i\n' % len(db))
279 statsfile.write(' Statistics last updated: %s\n\n' \
280 % time.strftime('%A, %B %d %Y, %H:%M:%S UTC', time.gmtime()))
283 if __name__ == '__main__':
284 main(sys.argv[1], sys.argv[2])