def ggplot2_options():
def normal_text():
return ggplot2.theme_text(family='serif', size=15)
def bold_text():
return ggplot2.theme_text(family='serif', face='bold', size=15)
def rotated_text():
return ggplot2.theme_text(family='serif',
face='bold',
size=15,
angle=90,
vjust=0.2)
return ggplot2.opts(
**{
'axis.title.x': ggplot2.theme_blank(),
'axis.title.y': rotated_text(),
'axis.text.x': normal_text(),
'axis.text.y': normal_text(),
'legend.title': bold_text(),
'legend.text': normal_text(),
'aspect.ratio': 0.6180339888,
'strip.text.x': normal_text(),
})
def interval(locus_table, interval_table, intervals, loci, boxplot = True):
qry = get_interval_query(intervals, loci, locus_table, interval_table)
frame = robjects.r('''data <- dbGetQuery(con, {})'''.format(qry))
# because we're sorting by interval, which is a factor, we need to
# explicitly re-sort the data by the first integer value
# of the interval. This is a bit cumbersome, because sorting
# in R is less than pleasant.
sort_string = '''data$interval <- factor(data$interval, {})'''.format(order_intervals(frame[1]))
robjects.r(sort_string)
gg_frame = ggplot2.ggplot(robjects.r('''data'''))
if boxplot:
plot = gg_frame + ggplot2.aes_string(x = 'interval', y = 'pi') + \
ggplot2.geom_boxplot(**{
'outlier.size':0,
'alpha':0.3
}
) + \
ggplot2.geom_jitter(ggplot2.aes_string(color = 'locus'), size = 3, \
alpha = 0.6, position=ggplot2.position_jitter(width=0.25)) + \
ggplot2.scale_y_continuous('phylogenetic informativeness') + \
ggplot2.scale_x_discrete('interval (years ago)')
else:
plot = gg_frame + ggplot2.aes_string(x = 'interval', y = 'pi',
fill='locus') + ggplot2.geom_bar() + \
ggplot2.facet_wrap(robjects.Formula('~ locus')) + \
ggplot2.opts(**{
'axis.text.x':ggplot2.theme_text(angle = -90, hjust = 0),
'legend.position':'none'
}) + \
ggplot2.scale_y_continuous('phylogenetic informativeness') + \
ggplot2.scale_x_discrete('interval (years ago)')
return plot
def single_locus_net_informativeness(locus_table, net_pi_table, locus):
qry = '''"SELECT {0}.locus, time, pi FROM {0}, {1}
WHERE {0}.id = {1}.id AND locus = '{2}'"'''.format(locus_table,
net_pi_table, locus)
frame = robjects.r('''dbGetQuery(con, {})'''.format(qry))
gg_frame = ggplot2.ggplot(frame)
plot = gg_frame + ggplot2.aes_string(x = 'time', y='pi') + \
ggplot2.geom_point(size = 3, alpha = 0.4) + \
ggplot2.scale_x_reverse('years ago') + \
ggplot2.scale_y_continuous('phylogenetic informativeness') + \
ggplot2.opts(title = locus)
return plot
def runBoruta():
base.load("Rcode/zscores.RData")
base.source('Z:/Cristina/MassNonmass/codeProject/codeBase/trainClassifier/Rcode/borutaRelevance.R')
outputBoruta = globalenv['findRelevant'](globalenv['massallfeatures'], globalenv['nonmassallfeatures'])
# generate boxplot comparison of relevant mass features vs. the same non-mass feature
plotgp = ggplot2.ggplot(outputBoruta.rx2("masszscore_selected")) + \
ggplot2.aes_string(x='MorN', y='zscores', fill = 'factor(MorN)') + \
ggplot2.geom_boxplot() + \
ggplot2.opts(title = "Comparison of Z-scores for Mass confirmed features", y="Z-scores")
plotgp.plot()
return
def plot(data, filename, title, ggplotter, xid="N", yid="RunTime", factorid="Step"):
df = make_dataframe(data, xid, yid, factorid)
grdevices.pdf(file=filename, width=10, height=6)
gp = ggplot2.ggplot(df)
pp = gp + \
ggplot2.aes_string(x=xid, y=yid) + \
ggplot2.aes_string(size=.5) + \
ggplotter() + \
ggplot2.aes_string(colour='factor(%s)' % factorid) + \
ggplot2.aes_string(fill='factor(%s)' % factorid) + \
ggplot2.opts(title=title) + \
ggplot2.scale_fill_brewer(palette="Set2") + \
ggplot2.scale_colour_brewer(palette="Set2")
pp.plot()
grdevices.dev_off()
def multiple_locus_net_informativeness_facet(locus_table, net_pi_table, loci):
if loci[0].lower() != 'all':
qry = '''"SELECT {0}.locus, time, pi FROM {0}, {1}
WHERE {0}.id = {1}.id and locus in {2}"'''.format(locus_table,
net_pi_table, tuple(loci))
else:
qry = '''"SELECT {0}.locus, time, pi FROM {0}, {1}
WHERE {0}.id = {1}.id"'''.format(locus_table,
net_pi_table)
frame = robjects.r('''dbGetQuery(con, {})'''.format(qry))
gg_frame = ggplot2.ggplot(frame)
plot = gg_frame + ggplot2.aes_string(x = 'time', y='pi') + \
ggplot2.geom_point(ggplot2.aes_string(colour = 'locus'), size = 3, \
alpha = 0.4) + ggplot2.scale_x_reverse('years ago') + \
ggplot2.facet_wrap(robjects.Formula('~ locus')) + \
ggplot2.opts(**{'legend.position' : 'none'}) + \
ggplot2.scale_y_continuous('phylogenetic informativeness')
return plot
def ggplot2_options ():
def normal_text():
return ggplot2.theme_text(family = 'serif', size = 15)
def bold_text():
return ggplot2.theme_text(family = 'serif', face = 'bold', size = 15)
def rotated_text():
return ggplot2.theme_text(family = 'serif', face = 'bold',
size = 15, angle=90, vjust=0.2)
return ggplot2.opts (**{'axis.title.x' : ggplot2.theme_blank(),
'axis.title.y' : rotated_text(),
'axis.text.x' : normal_text(),
'axis.text.y' : normal_text(),
'legend.title' : bold_text(),
'legend.text' : normal_text(),
'aspect.ratio' : 0.6180339888,
'strip.text.x' : normal_text(),
})
def ggplot2_options():
return ggplot2.opts(
**{
'axis.title.x':
ggplot2.theme_blank(),
'axis.title.y':
ggplot2.theme_text(
family='serif', face='bold', size=15, angle=90, vjust=0.2),
'axis.text.x':
ggplot2.theme_text(family='serif', size=15),
'axis.text.y':
ggplot2.theme_text(family='serif', size=15),
'legend.title':
ggplot2.theme_text(family='serif', face='bold', size=15),
'legend.text':
ggplot2.theme_text(family='serif', size=15),
'aspect.ratio':
0.6180339888,
})
def retrieve(request):
stockname = request.POST['stockname']
q = GoogleIntradayQuote(stockname, 300, 1)
qlines = q.to_csv().strip().split("\n")
if not qlines:
e = "Couldn't find ", stockname, " data"
return render(request, 'stocks/index.html', {'error_message': e})
else:
s = Stock(ticker=q.symbol,
start_date=timezone.now() - datetime.timedelta(days=1))
s.save()
for tick in q.get_ticks():
print "tick: ", tick,
dt = parse_datetime(tick[0])
p = tick[1]
print " price: ", p
s.tick_set.create(time=dt, price=p)
s.save()
# tells the R plotting device to write the plot to a file
f = "static/stocks/images/tmp.png"
grdevices.png(file=f, width=512, height=512)
# make a random plot
rnorm = stats.rnorm
df = {'value': rnorm(300, mean=0) + rnorm(100, mean=3),
'other_value': rnorm(300, mean=0) + rnorm(100, mean=3),
'mean': IntVector([0, ] * 300 + [3, ] * 100)}
dataf_rnorm = robjects.DataFrame(df)
gp = ggplot2.ggplot(dataf_rnorm)
pp = gp + \
ggplot2.aes_string(x='value', y='other_value') + \
ggplot2.geom_bin2d() + \
ggplot2.opts(title='geom_bin2d')
pp.plot()
grdevices.dev_off()
context = {'stockname': stockname,
'history': qlines}
return render(request, 'stocks/retrieve.html', context)
def mem_usage_graph(cfg):
r = robjects.r
varis = []
langs = []
probs = []
mems = []
for var in cfg.variations:
for lang in cfg.languages:
for prob in cfg.problems:
mem_filename = get_mem_output(lang, prob, var)
with open(mem_filename, 'r') as mem_file:
mem = mem_file.readline()
mems.append(float(mem))
varis.append(pretty_varis[var])
langs.append(pretty_langs[lang])
probs.append(prob)
# memory usage is a simple histogram with all information in one graph.
r.pdf('bargraph-memusage.pdf', height=pdf_height(), width=pdf_width())
df = robjects.DataFrame({
'Language': StrVector(langs),
'Problem': StrVector(probs),
'Variation': StrVector(varis),
'Mem': FloatVector(mems)
})
gp = ggplot2.ggplot(df)
# we rotate the x labels to make sure they don't overlap
pp = gp +\
ggplot2.opts (**{'axis.text.x': ggplot2.theme_text (angle = 90, hjust=1)}) + \
ggplot2.aes_string (x='Problem', y='Mem', fill='Language') + \
ggplot2.geom_bar (position='dodge', stat='identity') + \
ggplot2.facet_wrap ('Variation') + \
ggplot2_options () + \
ggplot2_colors () + \
robjects.r('scale_x_discrete(limits=c("randmat", "thresh", "winnow", "outer", "product", "chain"))') +\
robjects.r('ylab("Memory usage (in bytes)")')# + \
pp.plot()
r['dev.off']()
def mem_usage_graph (cfg):
r = robjects.r
varis = []
langs = []
probs = []
mems = []
for var in cfg.variations:
for lang in cfg.languages:
for prob in cfg.problems:
mem_filename = get_mem_output (lang, prob, var)
with open (mem_filename, 'r') as mem_file:
mem = mem_file.readline()
mems.append (float (mem))
varis.append (pretty_varis [var])
langs.append (pretty_langs [lang])
probs.append (prob)
# memory usage is a simple histogram with all information in one graph.
r.pdf ('bargraph-memusage.pdf', height=pdf_height (), width=pdf_width ())
df = robjects.DataFrame({'Language': StrVector (langs),
'Problem': StrVector (probs),
'Variation' : StrVector (varis),
'Mem' : FloatVector (mems)
})
gp = ggplot2.ggplot (df)
# we rotate the x labels to make sure they don't overlap
pp = gp +\
ggplot2.opts (**{'axis.text.x': ggplot2.theme_text (angle = 90, hjust=1)}) + \
ggplot2.aes_string (x='Problem', y='Mem', fill='Language') + \
ggplot2.geom_bar (position='dodge', stat='identity') + \
ggplot2.facet_wrap ('Variation') + \
ggplot2_options () + \
ggplot2_colors () + \
robjects.r('scale_x_discrete(limits=c("randmat", "thresh", "winnow", "outer", "product", "chain"))') +\
robjects.r('ylab("Memory usage (in bytes)")')# + \
pp.plot ()
r['dev.off']()
def line_plot (cfg, var, control, change_name, changing, selector, base_selector, basis):
speedups = []
thrds = []
changes = []
lowers = []
uppers = []
for n in cfg.threads:
probs.append ('ideal')
langs.append ('ideal')
speedups.append (n)
thrds.append (n)
changes.append ('ideal')
lowers.append (n)
uppers.append (n)
for c in changing:
sel = selector (c)
# sequential base
base = FloatVector (base_selector(c))
# base with p = 1
base_p1 = FloatVector (sel(1))
# use fastest sequential program
if basis == 'fastest' and mean (base_p1) < mean(base):
base = base_p1
elif basis == 'seq':
pass
elif basis == 'p1':
base = base_p1
for n in cfg.threads:
ntimes = FloatVector (sel(n))
# ratio confidence interval
labels = ['Base'] * r.length(base)[0] + ['N']*r.length (ntimes)[0]
df = DataFrame ({'Times': base + ntimes,
'Type': StrVector(labels)})
ratio_test = r['pairwiseCI'] (r('Times ~ Type'), data=df,
control='N',
method='Param.ratio',
**{'var.equal': False,
'conf.level': 0.999})[0][0]
lowers.append (ratio_test[1][0])
uppers.append (ratio_test[2][0])
mn = mean (ntimes)
speedups.append (mean(base) / mn)
# plot slowdowns
#speedups.append (-mn/base)#(base / mn)
thrds.append (n)
if change_name == 'Language':
changes.append (pretty_langs [c])
else:
changes.append (c)
df = DataFrame ({'Speedup': FloatVector (speedups),
'Threads': IntVector (thrds),
change_name: StrVector (changes),
'Lower': FloatVector (lowers),
'Upper': FloatVector (uppers)
})
ideal_changing = ['ideal']
if change_name == 'Language':
ideal_changing.extend ([pretty_langs [c] for c in changing])
else:
ideal_changing.extend (changing)
legendVec = IntVector (range (len (ideal_changing)))
legendVec.names = StrVector (ideal_changing)
gg = ggplot2.ggplot (df)
limits = ggplot2.aes (ymax = 'Upper', ymin = 'Lower')
dodge = ggplot2.position_dodge (width=0.9)
pp = gg + \
ggplot2.geom_line() + ggplot2.geom_point(size=3) +\
ggplot2.aes_string(x='Threads', y='Speedup',
group=change_name, color=change_name,
shape=change_name) + \
ggplot2.scale_shape_manual(values=legendVec) + \
ggplot2.geom_errorbar (limits, width=0.25) + \
ggplot2_options () + \
ggplot2_colors () + \
ggplot2.opts (**{'axis.title.x' : ggplot2.theme_text(family = 'serif', face = 'bold', size = 15, vjust=-0.2)}) + \
robjects.r('ylab("Speedup")') + \
robjects.r('xlab("Cores")')
# ggplot2.xlim (min(threads), max(threads)) + ggplot2.ylim(min(threads), max(threads)) +\
pp.plot()
r['dev.off']()
def as_dataframe (cfg, results, basis):
r = robjects.r
varis = []
langs = []
probs = []
times = []
threads = []
# speedups, with upper and lower bounds below
speedups = []
speedup_lowers = []
speedup_uppers = []
ses = [] # standard errors
mems = [] # memory usage
langs_ideal = list (cfg.languages)
langs_ideal.append ('ideal')
probs_ideal = list (cfg.problems)
probs_ideal.append ('ideal')
for var in cfg.variations:
for lang in langs_ideal: # cfg.languages:
for prob in probs_ideal: # cfg.problems:
for thread in cfg.threads:
if lang == 'ideal' and prob == 'ideal':
continue
elif lang == 'ideal' or prob == 'ideal':
varis.append (var)
langs.append (pretty_langs[lang])
probs.append (prob)
threads.append (thread)
speedups.append (thread)
speedup_lowers.append (thread)
speedup_uppers.append (thread)
times.append (0)
ses.append(0)
mems.append (0)
continue
varis.append (var) # pretty_varis [var])
langs.append (pretty_langs [lang])
probs.append (prob)
threads.append (thread)
if var.find('seq') >= 0:
thread = cfg.threads[-1]
vals = FloatVector (results[thread][prob][var][lang][0])
time = mean (vals)
times.append (time)
#
# time confidence interval
#
t_result = r['t.test'] (FloatVector(vals),
**{" conf.level": 0.999}).rx ('conf.int')[0]
ses.append ((t_result[1] - t_result[0])/2)
#
# memory usage
#
mem_filename = get_mem_output (lang, prob, var)
with open (mem_filename, 'r') as mem_file:
mem = mem_file.readline()
mems.append (float (mem))
# we include dummy data for the sequential case to avoid the
# speedup calculation below
if var.find('seq') >= 0:
speedups.append (1)
speedup_lowers.append (1)
speedup_uppers.append (1)
continue
#
# speedup values and confidence intervals
#
seq_vals = results[cfg.threads[-1]][prob][var.replace ('par', 'seq')][lang][0]
# sequential base
base = FloatVector (seq_vals)
# base with p = 1
base_p1 = FloatVector (results[1][prob][var][lang][0])
# use fastest sequential program
if basis == 'fastest' and mean (base_p1) < mean(base):
base = base_p1
elif basis == 'seq':
pass
elif basis == 'p1':
base = base_p1
labels = ['Base'] * r.length(base)[0] + ['N']*r.length (vals)[0]
df = DataFrame ({'Times': base + vals,
'Type': StrVector(labels)})
ratio_test = r['pairwiseCI'] (r('Times ~ Type'), data=df,
control='N',
method='Param.ratio',
**{'var.equal': False})[0][0]
speedups.append (mean(base) / time)
speedup_lowers.append (ratio_test[1][0])
speedup_uppers.append (ratio_test[2][0])
df = robjects.DataFrame({'Language': StrVector (langs),
'Problem': StrVector (probs),
'Variation' : StrVector (varis),
'Threads': IntVector (threads),
'Time': FloatVector (times),
'SE': FloatVector (ses),
'Speedup': FloatVector (speedups),
'SpeedupLower': FloatVector (speedup_lowers),
'SpeedupUpper': FloatVector (speedup_uppers),
'Mem' : FloatVector (mems)
})
r.assign ('df', df)
r ('save (df, file="performance.Rda")')
# reshape the data to make variation not a column itself, but a part of
# the other columns describe ie, time, speedup, etc.
#
# also, remove the 'ideal' problem as we don't want it in this plot.
df = r('''
redf = reshape (df,
timevar="Variation",
idvar = c("Language","Problem","Threads"),
direction="wide")
redf$Problem <- factor(redf$Problem, levels = c("randmat","thresh","winnow","outer","product","chain"))
redf[which(redf$Problem != "ideal"),]
''')
r.pdf ('speedup-expertpar-all.pdf',
height=6.5, width=10)
change_name = 'Language'
legendVec = IntVector (range (len (langs_ideal)))
legendVec.names = StrVector (langs_ideal)
gg = ggplot2.ggplot (df)
limits = ggplot2.aes (ymax = 'SpeedupUpper.expertpar', ymin = 'SpeedupLower.expertpar')
dodge = ggplot2.position_dodge (width=0.9)
pp = gg + \
ggplot2.geom_line() + ggplot2.geom_point(size=2.5) +\
robjects.r('scale_color_manual(values = c("#ffcb7e", "#1da06b", "#b94646", "#00368a", "#CCCCCC"))') +\
ggplot2.aes_string(x='Threads', y='Speedup.expertpar',
group=change_name, color=change_name,
shape=change_name) + \
ggplot2.geom_errorbar (limits, width=0.25) + \
ggplot2.opts (**{'axis.title.x' : ggplot2.theme_text(family = 'serif', face = 'bold', size = 10, vjust=-0.2),
'axis.title.y' : ggplot2.theme_text(family = 'serif', face = 'bold', size = 10, angle=90, vjust=0.2),
'axis.text.x' : ggplot2.theme_text(family = 'serif', size = 10),
'axis.text.y' : ggplot2.theme_text(family = 'serif', size = 10),
'legend.title' : ggplot2.theme_text(family = 'serif', face = 'bold', size = 10),
'legend.text' : ggplot2.theme_text(family = 'serif', size = 10),
'strip.text.x' : ggplot2.theme_text(family = 'serif', size = 10),
'aspect.ratio' : 1,
}) + \
robjects.r('ylab("Speedup")') + \
robjects.r('xlab("Number of cores")') + \
ggplot2.facet_wrap ('Problem', nrow = 2)
pp.plot()
r['dev.off']()
#-- ggplot2mtcars-end
grdevices.dev_off()
grdevices.png('../../_static/graphics_ggplot2geombin2d.png',
width = 1000, height = 350, antialias="subpixel", type="cairo")
grid.newpage()
grid.viewport(layout=grid.layout(1, 3)).push()
vp = grid.viewport(**{'layout.pos.col':1, 'layout.pos.row': 1})
#-- ggplot2geombin2d-begin
gp = ggplot2.ggplot(dataf_rnorm)
pp = gp + \
ggplot2.aes_string(x='value', y='other_value') + \
ggplot2.geom_bin2d() + \
ggplot2.opts(title = 'geom_bin2d')
pp.plot(vp = vp)
#-- ggplot2geombin2d-end
vp = grid.viewport(**{'layout.pos.col':2, 'layout.pos.row': 1})
#-- ggplot2geomdensity2d-begin
gp = ggplot2.ggplot(dataf_rnorm)
pp = gp + \
ggplot2.aes_string(x='value', y='other_value') + \
ggplot2.geom_density2d() + \
ggplot2.opts(title = 'geom_density2d')
pp.plot(vp = vp)
#-- ggplot2geomdensity2d-end
vp = grid.viewport(**{'layout.pos.col':3, 'layout.pos.row': 1})
from rpy2.robjects.lib import ggplot2
p = ggplot2.ggplot(dataf) + \
ggplot2.geom_line(ggplot2.aes_string(x="n_loop",
y="time",
colour="code")) + \
ggplot2.geom_point(ggplot2.aes_string(x="n_loop",
y="time",
colour="code")) + \
ggplot2.facet_wrap(Formula('~sequence')) + \
ggplot2.scale_y_continuous('running time') + \
ggplot2.scale_x_continuous('repeated n times', ) + \
ggplot2.xlim(0, max(n_loops)) + \
ggplot2.opts(title = "Benchmark (running time)")
from rpy2.robjects.packages import importr
grdevices = importr('grDevices')
grdevices.png('../../_static/benchmark_sum.png',
width = 712, height = 512)
p.plot()
grdevices.dev_off()
#base = importr("base")
stats = importr('stats')
nlme = importr("nlme")
fit = nlme.lmList(Formula('time ~ n_loop | group'), data = dataf,
na_action = stats.na_exclude)
def plot(self, fn, x='x', y='y', col=None, group=None, w=1100, h=800, size=2, smooth=True, point=True, jitter=False, boxplot=False, boxplot2=False, title=False, flip=False, se=False, density=False, line=False):
df=self.df
#import math, datetime
grdevices = importr('grDevices')
if not title:
title=fn.split("/")[-1]
grdevices.png(file=fn, width=w, height=h)
gp = ggplot2.ggplot(df)
pp = gp
if col and group:
pp+=ggplot2.aes_string(x=x, y=y,col=col,group=group)
elif col:
pp+=ggplot2.aes_string(x=x, y=y,col=col)
elif group:
pp+=ggplot2.aes_string(x=x, y=y,group=group)
else:
pp+=ggplot2.aes_string(x=x, y=y)
if boxplot:
if col:
pp+=ggplot2.geom_boxplot(ggplot2.aes_string(fill=col),color='blue')
else:
pp+=ggplot2.geom_boxplot(color='blue')
if point:
if jitter:
if col:
pp+=ggplot2.geom_point(ggplot2.aes_string(fill=col,col=col),size=size,position='jitter')
else:
pp+=ggplot2.geom_point(size=size,position='jitter')
else:
if col:
pp+=ggplot2.geom_point(ggplot2.aes_string(fill=col,col=col),size=size)
else:
pp+=ggplot2.geom_point(size=size)
if boxplot2:
if col:
pp+=ggplot2.geom_boxplot(ggplot2.aes_string(fill=col),color='blue',outlier_colour="NA")
else:
pp+=ggplot2.geom_boxplot(color='blue')
if smooth:
if smooth=='lm':
if col:
pp+=ggplot2.stat_smooth(ggplot2.aes_string(col=col),size=1,method='lm',se=se)
else:
pp+=ggplot2.stat_smooth(col='blue',size=1,method='lm',se=se)
else:
if col:
pp+=ggplot2.stat_smooth(ggplot2.aes_string(col=col),size=1,se=se)
else:
pp+=ggplot2.stat_smooth(col='blue',size=1,se=se)
if density:
pp+=ggplot2.geom_density(ggplot2.aes_string(x=x,y='..count..'))
if line:
pp+=ggplot2.geom_line(position='jitter')
pp+=ggplot2.opts(**{'title' : title, 'axis.text.x': ggplot2.theme_text(size=24), 'axis.text.y': ggplot2.theme_text(size=24,hjust=1)} )
#pp+=ggplot2.scale_colour_brewer(palette="Set1")
pp+=ggplot2.scale_colour_hue()
if flip:
pp+=ggplot2.coord_flip()
pp.plot()
grdevices.dev_off()
print ">> saved: "+fn
def as_dataframe(cfg, results, basis):
r = robjects.r
varis = []
langs = []
probs = []
times = []
threads = []
# speedups, with upper and lower bounds below
speedups = []
speedup_lowers = []
speedup_uppers = []
ses = [] # standard errors
mems = [] # memory usage
langs_ideal = list(cfg.languages)
langs_ideal.append('ideal')
probs_ideal = list(cfg.problems)
probs_ideal.append('ideal')
for var in cfg.variations:
for lang in langs_ideal: # cfg.languages:
for prob in probs_ideal: # cfg.problems:
for thread in cfg.threads:
if lang == 'ideal' and prob == 'ideal':
continue
elif lang == 'ideal' or prob == 'ideal':
varis.append(var)
langs.append(pretty_langs[lang])
probs.append(prob)
threads.append(thread)
speedups.append(thread)
speedup_lowers.append(thread)
speedup_uppers.append(thread)
times.append(0)
ses.append(0)
mems.append(0)
continue
varis.append(var) # pretty_varis [var])
langs.append(pretty_langs[lang])
probs.append(prob)
threads.append(thread)
if var.find('seq') >= 0:
thread = cfg.threads[-1]
vals = FloatVector(results[thread][prob][var][lang][0])
time = mean(vals)
times.append(time)
#
# time confidence interval
#
t_result = r['t.test'](FloatVector(vals), **{
" conf.level": 0.999
}).rx('conf.int')[0]
ses.append((t_result[1] - t_result[0]) / 2)
#
# memory usage
#
mem_filename = get_mem_output(lang, prob, var)
with open(mem_filename, 'r') as mem_file:
mem = mem_file.readline()
mems.append(float(mem))
# we include dummy data for the sequential case to avoid the
# speedup calculation below
if var.find('seq') >= 0:
speedups.append(1)
speedup_lowers.append(1)
speedup_uppers.append(1)
continue
#
# speedup values and confidence intervals
#
seq_vals = results[cfg.threads[-1]][prob][var.replace(
'par', 'seq')][lang][0]
# sequential base
base = FloatVector(seq_vals)
# base with p = 1
base_p1 = FloatVector(results[1][prob][var][lang][0])
# use fastest sequential program
if basis == 'fastest' and mean(base_p1) < mean(base):
base = base_p1
elif basis == 'seq':
pass
elif basis == 'p1':
base = base_p1
labels = ['Base'
] * r.length(base)[0] + ['N'] * r.length(vals)[0]
df = DataFrame({
'Times': base + vals,
'Type': StrVector(labels)
})
ratio_test = r['pairwiseCI'](r('Times ~ Type'),
data=df,
control='N',
method='Param.ratio',
**{
'var.equal': False
})[0][0]
speedups.append(mean(base) / time)
speedup_lowers.append(ratio_test[1][0])
speedup_uppers.append(ratio_test[2][0])
df = robjects.DataFrame({
'Language': StrVector(langs),
'Problem': StrVector(probs),
'Variation': StrVector(varis),
'Threads': IntVector(threads),
'Time': FloatVector(times),
'SE': FloatVector(ses),
'Speedup': FloatVector(speedups),
'SpeedupLower': FloatVector(speedup_lowers),
'SpeedupUpper': FloatVector(speedup_uppers),
'Mem': FloatVector(mems)
})
r.assign('df', df)
r('save (df, file="performance.Rda")')
# reshape the data to make variation not a column itself, but a part of
# the other columns describe ie, time, speedup, etc.
#
# also, remove the 'ideal' problem as we don't want it in this plot.
df = r('''
redf = reshape (df,
timevar="Variation",
idvar = c("Language","Problem","Threads"),
direction="wide")
redf$Problem <- factor(redf$Problem, levels = c("randmat","thresh","winnow","outer","product","chain"))
redf[which(redf$Problem != "ideal"),]
''')
r.pdf('speedup-expertpar-all.pdf', height=6.5, width=10)
change_name = 'Language'
legendVec = IntVector(range(len(langs_ideal)))
legendVec.names = StrVector(langs_ideal)
gg = ggplot2.ggplot(df)
limits = ggplot2.aes(ymax='SpeedupUpper.expertpar',
ymin='SpeedupLower.expertpar')
dodge = ggplot2.position_dodge(width=0.9)
pp = gg + \
ggplot2.geom_line() + ggplot2.geom_point(size=2.5) +\
robjects.r('scale_color_manual(values = c("#ffcb7e", "#1da06b", "#b94646", "#00368a", "#CCCCCC"))') +\
ggplot2.aes_string(x='Threads', y='Speedup.expertpar',
group=change_name, color=change_name,
shape=change_name) + \
ggplot2.geom_errorbar (limits, width=0.25) + \
ggplot2.opts (**{'axis.title.x' : ggplot2.theme_text(family = 'serif', face = 'bold', size = 10, vjust=-0.2),
'axis.title.y' : ggplot2.theme_text(family = 'serif', face = 'bold', size = 10, angle=90, vjust=0.2),
'axis.text.x' : ggplot2.theme_text(family = 'serif', size = 10),
'axis.text.y' : ggplot2.theme_text(family = 'serif', size = 10),
'legend.title' : ggplot2.theme_text(family = 'serif', face = 'bold', size = 10),
'legend.text' : ggplot2.theme_text(family = 'serif', size = 10),
'strip.text.x' : ggplot2.theme_text(family = 'serif', size = 10),
'aspect.ratio' : 1,
}) + \
robjects.r('ylab("Speedup")') + \
robjects.r('xlab("Number of cores")') + \
ggplot2.facet_wrap ('Problem', nrow = 2)
pp.plot()
r['dev.off']()
def plot(self,
fn,
x='x',
y='y',
col=None,
group=None,
w=1100,
h=800,
size=2,
smooth=True,
point=True,
jitter=False,
boxplot=False,
boxplot2=False,
title=False,
flip=False,
se=False,
density=False,
line=False):
df = self.df
#import math, datetime
grdevices = importr('grDevices')
if not title:
title = fn.split("/")[-1]
grdevices.png(file=fn, width=w, height=h)
gp = ggplot2.ggplot(df)
pp = gp
if col and group:
pp += ggplot2.aes_string(x=x, y=y, col=col, group=group)
elif col:
pp += ggplot2.aes_string(x=x, y=y, col=col)
elif group:
pp += ggplot2.aes_string(x=x, y=y, group=group)
else:
pp += ggplot2.aes_string(x=x, y=y)
if boxplot:
if col:
pp += ggplot2.geom_boxplot(ggplot2.aes_string(fill=col),
color='blue')
else:
pp += ggplot2.geom_boxplot(color='blue')
if point:
if jitter:
if col:
pp += ggplot2.geom_point(ggplot2.aes_string(fill=col,
col=col),
size=size,
position='jitter')
else:
pp += ggplot2.geom_point(size=size, position='jitter')
else:
if col:
pp += ggplot2.geom_point(ggplot2.aes_string(fill=col,
col=col),
size=size)
else:
pp += ggplot2.geom_point(size=size)
if boxplot2:
if col:
pp += ggplot2.geom_boxplot(ggplot2.aes_string(fill=col),
color='blue',
outlier_colour="NA")
else:
pp += ggplot2.geom_boxplot(color='blue')
if smooth:
if smooth == 'lm':
if col:
pp += ggplot2.stat_smooth(ggplot2.aes_string(col=col),
size=1,
method='lm',
se=se)
else:
pp += ggplot2.stat_smooth(col='blue',
size=1,
method='lm',
se=se)
else:
if col:
pp += ggplot2.stat_smooth(ggplot2.aes_string(col=col),
size=1,
se=se)
else:
pp += ggplot2.stat_smooth(col='blue', size=1, se=se)
if density:
pp += ggplot2.geom_density(ggplot2.aes_string(x=x, y='..count..'))
if line:
pp += ggplot2.geom_line(position='jitter')
pp += ggplot2.opts(
**{
'title': title,
'axis.text.x': ggplot2.theme_text(size=24),
'axis.text.y': ggplot2.theme_text(size=24, hjust=1)
})
#pp+=ggplot2.scale_colour_brewer(palette="Set1")
pp += ggplot2.scale_colour_hue()
if flip:
pp += ggplot2.coord_flip()
pp.plot()
grdevices.dev_off()
print ">> saved: " + fn
width=1300,
height=1000)
## plot the map
## note that the order matters when we add another layer in ggplot (here IL_railroads): first aes, then data, that's different from R
## (see http://permalink.gmane.org/gmane.comp.python.rpy/2349)
## note that we use dictionary to set the opts to be able to set options as keywords, for example legend.key.size
p_map = ggplot2.ggplot(IL_final) + \
ggplot2.geom_polygon(ggplot2.aes(x = 'long', y = 'lat', group = 'group', color = 'ObamaShare', fill = 'ObamaShare')) + \
ggplot2.scale_fill_gradient(high = 'blue', low = 'red') + \
ggplot2.scale_fill_continuous(name = "Obama Vote Share") + \
ggplot2.scale_colour_continuous(name = "Obama Vote Share") + \
ggplot2.opts(**{'legend.position': 'left', 'legend.key.size': robjects.r.unit(2, 'lines'), 'legend.title' : ggplot2.theme_text(size = 14, hjust=0), \
'legend.text': ggplot2.theme_text(size = 12), 'title' : "Obama Vote Share and Distance to Railroads in IL", \
'plot.title': ggplot2.theme_text(size = 24), 'plot.margin': robjects.r.unit(robjects.r.rep(0,4),'lines'), \
'panel.background': ggplot2.theme_blank(), 'panel.grid.minor': ggplot2.theme_blank(), 'panel.grid.major': ggplot2.theme_blank(), \
'axis.ticks': ggplot2.theme_blank(), 'axis.title.x': ggplot2.theme_blank(), 'axis.title.y': ggplot2.theme_blank(), \
'axis.title.x': ggplot2.theme_blank(), 'axis.title.x': ggplot2.theme_blank(), 'axis.text.x': ggplot2.theme_blank(), \
'axis.text.y': ggplot2.theme_blank()} ) + \
ggplot2.geom_line(ggplot2.aes(x='long', y='lat', group='group'), data=IL_railroads, color='grey', size=0.2) + \
ggplot2.coord_equal()
p_map.plot()
## add the scatterplot
## define layout of subplot with viewports
vp_sub = grid.viewport(x=0.19, y=0.2, width=0.32, height=0.4)
p_sub = ggplot2.ggplot(RR_distance) + \
ggplot2.aes_string(x = 'OBAMA_SHAR', y= 'NEAR_DIST') + \
# print str(a)
try:
if dsumFC.has_key(drug):
dsumFC[drug]['Fold_Change'].append(math.log10(float(val)))
dsumY[drug]['Year'].append(yr)
else:
dsumFC[drug]= {'Fold_Change': [math.log10(float(val)),]}
dsumY[drug]= {'Year': [yr,]}
except:
print "FAILURE: dsumFC="+str(dsumFC)+"\n\ndsumY="+str(dsumY)
sys.exit()
drugs = dsumFC.keys()
for x in drugs:
od = rlc.OrdDict([('Fold_Change',robjects.FloatVector(dsumFC[x]['Fold_Change'])),('Year',robjects.FactorVector(dsumY[x]['Year'])),('Drug',robjects.FactorVector(x))])
grdevices.pdf(file="drugs.pdf",width=7,height=7)
dataf = robjects.DataFrame(od)
gp3 = ggplot2.ggplot(dataf)
pp3 = gp3 + ggplot2.scale_fill_brewer(palette='BrBG',name="Year")+ ggplot2.aes_string(x='Year',y='Fold_Change',fill='factor(Year)') + ggplot2.geom_boxplot() + ggplot2.opts(title = x+" Yearly Trend")
# pp3 = gp3 + ggplot2.scale_colour_hue(h=base.c(180,270),name="Year")+ ggplot2.aes_string(x='Year',y='Fold_Change',fill='factor(Year)') + ggplot2.geom_boxplot() + ggplot2.opts(title = x+" Yearly Trend")
#+ ggplot2.scale_y_log10()
pp3.plot()
grdevices.dev_off()
f.close()
print "\nfinished\n"
grdevices = importr('grDevices')
grdevices.png(file='/Users/user/Downloads/data/mapplot.png', width=1300, height=1000)
## plot the map
## note that the order matters when we add another layer in ggplot (here IL_railroads): first aes, then data, that's different from R
## (see http://permalink.gmane.org/gmane.comp.python.rpy/2349)
## note that we use dictionary to set the opts to be able to set options as keywords, for example legend.key.size
p_map = ggplot2.ggplot(IL_final) + \
ggplot2.geom_polygon(ggplot2.aes(x = 'long', y = 'lat', group = 'group', color = 'ObamaShare', fill = 'ObamaShare')) + \
ggplot2.scale_fill_gradient(high = 'blue', low = 'red') + \
ggplot2.scale_fill_continuous(name = "Obama Vote Share") + \
ggplot2.scale_colour_continuous(name = "Obama Vote Share") + \
ggplot2.opts(**{'legend.position': 'left', 'legend.key.size': robjects.r.unit(2, 'lines'), 'legend.title' : ggplot2.theme_text(size = 14, hjust=0), \
'legend.text': ggplot2.theme_text(size = 12), 'title' : "Obama Vote Share and Distance to Railroads in IL", \
'plot.title': ggplot2.theme_text(size = 24), 'plot.margin': robjects.r.unit(robjects.r.rep(0,4),'lines'), \
'panel.background': ggplot2.theme_blank(), 'panel.grid.minor': ggplot2.theme_blank(), 'panel.grid.major': ggplot2.theme_blank(), \
'axis.ticks': ggplot2.theme_blank(), 'axis.title.x': ggplot2.theme_blank(), 'axis.title.y': ggplot2.theme_blank(), \
'axis.title.x': ggplot2.theme_blank(), 'axis.title.x': ggplot2.theme_blank(), 'axis.text.x': ggplot2.theme_blank(), \
'axis.text.y': ggplot2.theme_blank()} ) + \
ggplot2.geom_line(ggplot2.aes(x='long', y='lat', group='group'), data=IL_railroads, color='grey', size=0.2) + \
ggplot2.coord_equal()
p_map.plot()
## add the scatterplot
## define layout of subplot with viewports
vp_sub = grid.viewport(x = 0.19, y = 0.2, width = 0.32, height = 0.4)
p_sub = ggplot2.ggplot(RR_distance) + \
ggplot2.aes_string(x = 'OBAMA_SHAR', y= 'NEAR_DIST') + \
[d['code'][x] + ':' + d['sequence'][x] for x in xrange(len(d['n_loop']))])
dataf = DataFrame(d)
from rpy2.robjects.lib import ggplot2
p = ggplot2.ggplot(dataf) + \
ggplot2.geom_line(ggplot2.aes_string(x="n_loop",
y="time",
colour="code")) + \
ggplot2.geom_point(ggplot2.aes_string(x="n_loop",
y="time",
colour="code")) + \
ggplot2.facet_wrap(Formula('~sequence')) + \
ggplot2.scale_y_continuous('running time') + \
ggplot2.scale_x_continuous('repeated n times', ) + \
ggplot2.xlim(0, max(n_loops)) + \
ggplot2.opts(title = "Benchmark (running time)")
from rpy2.robjects.packages import importr
grdevices = importr('grDevices')
grdevices.png('../../_static/benchmark_sum.png', width=712, height=512)
p.plot()
grdevices.dev_off()
#base = importr("base")
stats = importr('stats')
nlme = importr("nlme")
fit = nlme.lmList(Formula('time ~ n_loop | group'),
data=dataf,
na_action=stats.na_exclude)
# scale to R's slope
def line_plot(cfg, var, control, change_name, changing, selector,
base_selector, basis):
speedups = []
thrds = []
changes = []
lowers = []
uppers = []
for n in cfg.threads:
probs.append('ideal')
langs.append('ideal')
speedups.append(n)
thrds.append(n)
changes.append('ideal')
lowers.append(n)
uppers.append(n)
for c in changing:
sel = selector(c)
# sequential base
base = FloatVector(base_selector(c))
# base with p = 1
base_p1 = FloatVector(sel(1))
# use fastest sequential program
if basis == 'fastest' and mean(base_p1) < mean(base):
base = base_p1
elif basis == 'seq':
pass
elif basis == 'p1':
base = base_p1
for n in cfg.threads:
ntimes = FloatVector(sel(n))
# ratio confidence interval
labels = ['Base'] * r.length(base)[0] + ['N'] * r.length(ntimes)[0]
df = DataFrame({'Times': base + ntimes, 'Type': StrVector(labels)})
ratio_test = r['pairwiseCI'](r('Times ~ Type'),
data=df,
control='N',
method='Param.ratio',
**{
'var.equal': False,
'conf.level': 0.999
})[0][0]
lowers.append(ratio_test[1][0])
uppers.append(ratio_test[2][0])
mn = mean(ntimes)
speedups.append(mean(base) / mn)
# plot slowdowns
#speedups.append (-mn/base)#(base / mn)
thrds.append(n)
if change_name == 'Language':
changes.append(pretty_langs[c])
else:
changes.append(c)
df = DataFrame({
'Speedup': FloatVector(speedups),
'Threads': IntVector(thrds),
change_name: StrVector(changes),
'Lower': FloatVector(lowers),
'Upper': FloatVector(uppers)
})
ideal_changing = ['ideal']
if change_name == 'Language':
ideal_changing.extend([pretty_langs[c] for c in changing])
else:
ideal_changing.extend(changing)
legendVec = IntVector(range(len(ideal_changing)))
legendVec.names = StrVector(ideal_changing)
gg = ggplot2.ggplot(df)
limits = ggplot2.aes(ymax='Upper', ymin='Lower')
dodge = ggplot2.position_dodge(width=0.9)
pp = gg + \
ggplot2.geom_line() + ggplot2.geom_point(size=3) +\
ggplot2.aes_string(x='Threads', y='Speedup',
group=change_name, color=change_name,
shape=change_name) + \
ggplot2.scale_shape_manual(values=legendVec) + \
ggplot2.geom_errorbar (limits, width=0.25) + \
ggplot2_options () + \
ggplot2_colors () + \
ggplot2.opts (**{'axis.title.x' : ggplot2.theme_text(family = 'serif', face = 'bold', size = 15, vjust=-0.2)}) + \
robjects.r('ylab("Speedup")') + \
robjects.r('xlab("Cores")')
# ggplot2.xlim (min(threads), max(threads)) + ggplot2.ylim(min(threads), max(threads)) +\
pp.plot()
r['dev.off']()