def make_color_panel(df):
"""Map categorical values to colors"""
import brewer2mpl
set1 = brewer2mpl.get_map('Set1','Qualitative',9)
dark2 = brewer2mpl.get_map('Dark2','Qualitative',8)
out = {}
if 'condition' in df.columns:
d = {'MUT' : set1.mpl_colors[0], 'WT' : set1.mpl_colors[1]}
out['condition'] = pd.DataFrame([ d[x] for x in df.condition ],
index=df.index)
if 'disease_abbr' in df.columns:
d = {'BRCA' : dark2.mpl_colors[0], 'CLL' : dark2.mpl_colors[2],
'LUAD' : dark2.mpl_colors[1], 'LUSC' : dark2.mpl_colors[3],
'UM' : dark2.mpl_colors[-2]}
out['disease_abbr'] = pd.DataFrame([ d[x] for x in df.disease_abbr ],
index=df.index)
if 'HEAT 5-9' in df.columns:
d = {True:dark2.mpl_colors[-3], False:(1,1,1)}
out['HEAT 5-9'] = pd.DataFrame([ d[x] for x in df['HEAT 5-9']],
index=df.index)
if 'IGHV' in df.columns:
d = {'mutated':set1.mpl_colors[2], 'unmutated':set1.mpl_colors[3]}
out['IGHV'] = pd.DataFrame([ d[x] for x in df['IGHV']],
index=df.index)
if 'ZAP70' in df.columns:
d = {'positive':set1.mpl_colors[4], 'negative':set1.mpl_colors[6]}
out['ZAP70'] = pd.DataFrame([ d[x] for x in df['ZAP70']],
index=df.index)
return pd.Panel(out)
def plot(self):
# Data #
self.frame = self.parent.taxa_table.apply(lambda x: 100*x/x.sum(), axis=1)
# Take only the ones we use #
self.samples = [s for s in self.parent.samples if s.short_name in self.parent.taxa_table.index]
# Maybe drop tributaries or main river #
if self.tributary: self.samples = [s for s in self.samples if s.info['Tributary']==self.tributary]
# Sort #
key = lambda s: (-int(s.info['Tributary']), s.info['Filter_fraction'], s.short_name)
self.samples = sorted(self.samples, key=key)
self.frame = self.frame.reindex(index=[s.short_name for s in self.samples])
# Colors #
colors = brewer2mpl.get_map('Set1', 'qualitative', 8).mpl_colors
colors.reverse()
colors += brewer2mpl.get_map('Set2', 'qualitative', 8).mpl_colors
colors += brewer2mpl.get_map('Set3', 'qualitative', 8).mpl_colors
# Plot #
fig = pyplot.figure()
axes = self.frame.plot(kind='bar', stacked=True, color=colors)
fig = pyplot.gcf()
# Other #
axes.set_title('Species relative abundances per sample (blasting against "%s" database)' % self.parent.taxonomy.database)
axes.set_ylabel('Relative abundances in percent')
axes.xaxis.grid(False)
axes.yaxis.grid(False)
axes.set_ylim([0,100])
# Put a legend below current axis
axes.legend(loc='upper center', bbox_to_anchor=(0.5, -0.10), fancybox=True, shadow=True, ncol=5)
# Save it #
self.save_plot(fig, axes, width=24.0, height=14.0, bottom=0.20, top=0.97, left=0.04, right=0.98)
pyplot.close(fig)
def make_ssps(targetSN, w1, w2):
""" Make stellar kinematics"""
tables = []
for i in range(len(targetSN)):
tables.append("ssps_sn{0}_w{1}_{2}_TMJ10ext.txt".format(targetSN[i],
w1, w2))
columns = np.array([1,4,7,10])
# Calculating limits
lims = []
pers = [15, 20, 2, 15]
for j,i in enumerate(columns):
files = [os.path.join(data_dir, "combined_{0}".format(field), tab)
for tab, field in zip(tables, fields)]
a = np.concatenate([np.loadtxt(fname, usecols=(i,)) for fname in files])
q1, q2 = np.percentile(a[np.isfinite(a)], [pers[j], 100 - pers[j]])
lims.append([q1, q2])
# lims[0] = [8, 11]
# lims[1] = [-0.45, 0.36]
# lims[2] = [-0.08, 0.42]
# lims[3] = [-0.71, 0.10]
cmaps = [brewer2mpl.get_map('Reds', 'sequential', 6).mpl_colormap,
brewer2mpl.get_map('Blues', 'sequential', 6).mpl_colormap,
brewer2mpl.get_map('Greens', 'sequential', 6).mpl_colormap,
brewer2mpl.get_map('Oranges', 'sequential', 6).mpl_colormap]
labels = ["Age (Gyr)", "[Z/H]", "[$\\alpha$/Fe]", "[Fe/H]"]
cb_fmts = ["%.1f", "%.2f", "%.2f", "%.2f"]
yloss = [0.6, 0.6, 0.6, 0.6]
names = ["age", "metal", "alpha", "iron"]
plot(targetSN, 4500, 5500, tables, names, columns, lims, cmaps, labels,
cb_fmts, yloss, cbbox="regular")
return
def test_get_map_reverse():
bmap = brewer2mpl.get_map('Greens', 'Sequential', 8)
bmap_r = brewer2mpl.get_map('Greens', 'Sequential', 8, reverse=True)
assert bmap.type == bmap_r.type
assert bmap.name + '_r' == bmap_r.name
assert bmap.number == bmap_r.number
assert bmap.colors == reverse(bmap_r.colors)
def prettyplot():
import brewer2mpl
# Get Set2 from ColorBrewer, a set of colors deemed colorblind-safe and
# pleasant to look at by Drs. Cynthia Brewer and Mark Harrower of Pennsylvania
# State University. These colors look lovely together, and are less
# saturated than those colors in Set1. For more on ColorBrewer, see:
# - Flash-based interactive map:
# http://colorbrewer2.org/
# - A quick visual reference to every ColorBrewer scale:
# http://bl.ocks.org/mbostock/5577023
set2 = brewer2mpl.get_map("Set2", "qualitative", 8).mpl_colors
# Another ColorBrewer scale. This one has nice "traditional" colors like
# reds and blues
set1 = brewer2mpl.get_map("Set1", "qualitative", 9).mpl_colors
# Set some commonly used colors
almost_black = "#262626"
light_grey = np.array([float(248) / float(255)] * 3)
reds = mpl.cm.Reds
reds.set_bad("white")
reds.set_under("white")
blues_r = mpl.cm.Blues_r
blues_r.set_bad("white")
blues_r.set_under("white")
# Need to 'reverse' red to blue so that blue=cold=small numbers,
# and red=hot=large numbers with '_r' suffix
blue_red = brewer2mpl.get_map("RdBu", "Diverging", 11, reverse=True).mpl_colormap
green_purple = brewer2mpl.get_map("PRGn", "diverging", 11).mpl_colormap
red_purple = brewer2mpl.get_map("RdPu", "Sequential", 9).mpl_colormap
# Default "patches" like scatterplots
mpl.rcParams["patch.linewidth"] = 0.75 # edge width in points
# Default empty circle with a colored outline
mpl.rcParams["patch.facecolor"] = "none"
mpl.rcParams["patch.edgecolor"] = set2[0]
# Change the default axis colors from black to a slightly lighter black,
# and a little thinner (0.5 instead of 1)
mpl.rcParams["axes.edgecolor"] = almost_black
mpl.rcParams["axes.labelcolor"] = almost_black
mpl.rcParams["axes.linewidth"] = 0.5
# Make the default grid be white so it "removes" lines rather than adds
mpl.rcParams["grid.color"] = "white"
# change the tick colors also to the almost black
mpl.rcParams["ytick.color"] = almost_black
mpl.rcParams["xtick.color"] = almost_black
# change the text colors also to the almost black
mpl.rcParams["text.color"] = almost_black
return almost_black, blues_r, reds, blue_red, set1, set2, light_grey, green_purple, red_purple
def get_color_map(name, map_type, number, reverse=False, get_map=False):
"""
Parameters
----------
name : str
Name of color map. Use `print_maps` to see available color maps.
map_type : {'Sequential', 'Diverging', 'Qualitative'}
Select color map type.
number : int
Number of defined colors in color map.
reverse : bool, optional
Set to True to get the reversed color map.
All colors:
Sequential
Blues : {3, 4, 5, 6, 7, 8, 9}
BuGn : {3, 4, 5, 6, 7, 8, 9}
BuPu : {3, 4, 5, 6, 7, 8, 9}
GnBu : {3, 4, 5, 6, 7, 8, 9}
Greens : {3, 4, 5, 6, 7, 8, 9}
Greys : {3, 4, 5, 6, 7, 8, 9}
OrRd : {3, 4, 5, 6, 7, 8, 9}
Oranges : {3, 4, 5, 6, 7, 8, 9}
PuBu : {3, 4, 5, 6, 7, 8, 9}
PuBuGn : {3, 4, 5, 6, 7, 8, 9}
PuRd : {3, 4, 5, 6, 7, 8, 9}
Purples : {3, 4, 5, 6, 7, 8, 9}
RdPu : {3, 4, 5, 6, 7, 8, 9}
Reds : {3, 4, 5, 6, 7, 8, 9}
YlGn : {3, 4, 5, 6, 7, 8, 9}
YlGnBu : {3, 4, 5, 6, 7, 8, 9}
YlOrBr : {3, 4, 5, 6, 7, 8, 9}
YlOrRd : {3, 4, 5, 6, 7, 8, 9}
Diverging
BrBG : {3, 4, 5, 6, 7, 8, 9, 10, 11}
PRGn : {3, 4, 5, 6, 7, 8, 9, 10, 11}
PiYG : {3, 4, 5, 6, 7, 8, 9, 10, 11}
PuOr : {3, 4, 5, 6, 7, 8, 9, 10, 11}
RdBu : {3, 4, 5, 6, 7, 8, 9, 10, 11}
RdGy : {3, 4, 5, 6, 7, 8, 9, 10, 11}
RdYlBu : {3, 4, 5, 6, 7, 8, 9, 10, 11}
RdYlGn : {3, 4, 5, 6, 7, 8, 9, 10, 11}
Spectral : {3, 4, 5, 6, 7, 8, 9, 10, 11}
Qualitative
Accent : {3, 4, 5, 6, 7, 8}
Dark2 : {3, 4, 5, 6, 7, 8}
Paired : {3, 4, 5, 6, 7, 8, 9, 10, 11, 12}
Pastel1 : {3, 4, 5, 6, 7, 8, 9}
Pastel2 : {3, 4, 5, 6, 7, 8}
Set1 : {3, 4, 5, 6, 7, 8, 9}
Set2 : {3, 4, 5, 6, 7, 8}
Set3 : {3, 4, 5, 6, 7, 8, 9, 10, 11, 12}
"""
if get_map is True:
return(brewer2mpl.get_map(name, map_type, number).mpl_colormap)
else:
return(brewer2mpl.get_map(name, map_type, number).mpl_colors)
def _register_color_lists(self, length=default_lst):
cc = mpl.colors.ColorConverter()
self._color_lists['matplotlib'] = ['b', 'g', 'r', 'c', 'm', 'y', 'k']
self._color_lists['simple'] = ['#FF0000', '#FFD300', '#3914AF', '#00CC00']
self._color_lists['brewer'] = ['#e66101', '#fdb863', '#b2abd2', '#5e3c99']
for i in self._color_lists:
self._color_lists[i] = [cc.to_rgb(item) for item in self._color_lists[i]]
self._color_lists['cblind'] = [(0.9,0.6,0.0), (0.35,0.70,0.90), (0.00,0.60,0.50), (0.00,0.45,0.70), (0.80,0.40,0.00), (0.80,0.60,0.70)]
self._color_lists['brewer1'] = brewer2mpl.get_map('Set1', 'qualitative', length).mpl_colors
self._color_lists['brewer2'] = brewer2mpl.get_map('Set2', 'qualitative', length).mpl_colors
def __init__(self, colormap, alpha = 1):
self.colormap = colormap
self.set2 = colormap.mpl_colors
self.alpha = alpha
# Another ColorBrewer scale. This one has nice "traditional" colors like
# reds and blues
self.set1 = brewer2mpl.get_map('Set1', 'qualitative', 9).mpl_colors
mpl.rcParams['axes.color_cycle'] = self.set2
# Set some commonly used colors
self.almost_black = '#262626'
self.light_grey = np.array([float(248) / float(255)] * 3)
reds = mpl.cm.Reds
reds.set_bad('white')
reds.set_under('white')
blues_r = mpl.cm.Blues_r
blues_r.set_bad('white')
blues_r.set_under('white')
# Need to 'reverse' red to blue so that blue=cold=small numbers,
# and red=hot=large numbers with '_r' suffix
self.blue_red = brewer2mpl.get_map('RdBu', 'Diverging', 11,
reverse=True).mpl_colormap
# Default "patches" like scatterplots
mpl.rcParams['patch.linewidth'] = 0.75 # edge width in points
# Default empty circle with a colored outline
mpl.rcParams['patch.facecolor'] = 'none'
mpl.rcParams['patch.edgecolor'] = self.set2[0]
# Change the default axis colors from black to a slightly lighter black,
# and a little thinner (0.5 instead of 1)
mpl.rcParams['axes.edgecolor'] = self.almost_black
mpl.rcParams['axes.labelcolor'] = self.almost_black
mpl.rcParams['axes.linewidth'] = 0.5
# Make the default grid be white so it "removes" lines rather than adds
mpl.rcParams['grid.color'] = 'white'
# change the tick colors also to the almost black
mpl.rcParams['ytick.color'] = self.almost_black
mpl.rcParams['xtick.color'] = self.almost_black
# change the text colors also to the almost black
mpl.rcParams['text.color'] = self.almost_black
plt.rc('text', usetex=True)
plt.rc('font', family='serif')
plt.rc('font', size='16')
def colors(tables):
try:
import brewer2mpl
allcolors1=brewer2mpl.get_map('Set1', 'Qualitative',7).mpl_colors
allcolors2=brewer2mpl.get_map('Set2', 'Qualitative',6).mpl_colors
allcolors=allcolors1+allcolors2
except:
allcolors=['b','g','r','c','m','y','k','w']
while len(tables)> len(allcolors):
allcolors+=allcolors
return allcolors[:len(tables)]
def do_for_dataset(data, data_name):
transformed_data, pca, components = calculate_pca(data, n_components=3, normalize = True)
colormap = brewer2mpl.get_map('RdBu', 'diverging', 4, reverse=True)
filename = figure_save_path + dataset + '_issue_26_1_%s_PCA_3components.png'%data_name
print "Making scatter plot of PCA decompositions of %s data for dataset %s"%(data_name, dataset)
make_color_grouped_scatter_plot(data_frame=transformed_data, x_name='d1', y_name='d2', color_by='d3', filename=filename, colormap=colormap)
kmeans = KMeans(n_clusters = n_clusters)
kmeans.fit(transformed_data.values)
colormap = brewer2mpl.get_map('Set2', 'Qualitative', n_clusters, reverse=True)
filename = figure_save_path + dataset + '_issue_26_2_%s_clusters_in_PCA_space.png'%data_name
print "Making scatter plot of K-means clusters of %s data for dataset %s"%(data_name, dataset)
make_scatter_plot_for_labelled_data(data_frame=transformed_data, x_name='d1', y_name='d2', labels=kmeans.labels_, filename=filename, colormap = colormap, legend=True)
def _colourise_heatmap(self,colourscheme,maxColours):
""" by this stage we already have a regions_merged structure with the depth. This method turns those depths into RGB calls using colorbrewer
"""
## what's the maximum depth?
maxRec = max( [ max( [int(y['depth']) for y in self.regions_merged[taxa] ] ) for taxa in self.regions_merged ] )
## what scheme are we employing?
if colourscheme in ['BrBG','PRGn','PiYG','PuOr','RdBu','RdGy','RdYlBu','RdYlGn','Spectral']:
category="Diverging"
maxCol = 11
reverse = True
elif colourscheme in ['Blues','BuGn','BuPu','GnBu','Greens','Greys','OrRd','Oranges','PuBu','PuBuGn','PuRd','Purples','RdPu','Reds','YlGn','YlGnBu','YlOrBr','YlOrRd']:
category="Sequential"
maxCol = 9
reverse = False
else:
print "Invalid ColorBrewer scheme. Tabfile will not have any colours."
return 1
if maxColours < maxCol: maxCol = maxColours
if maxRec < 3: maxRec = 3 ## brewer can't handle only 2
# create depth -> RGB maps
depth2RGB = {}
if maxRec <= maxCol: ## easy! 1-1 map
print "Colorising heatmap using the Colour Brewer scheme {} with {} colours for {} levels".format(colourscheme,maxRec,maxRec)
brewerMap = brewer2mpl.get_map(colourscheme,category,maxRec,reverse=reverse).colors
for x in xrange(1,maxRec+1):
depth2RGB[x] = brewerMap[x-1]
else:
print "Colorising heatmap using the Colour Brewer scheme {} with {} colours for {} levels by binning all recombinations >= {} into the top colour".format(colourscheme,maxCol,maxRec,maxCol)
## we will just use the max num of categories here!
levels_per_colour = int(ceil(float(maxRec)/maxCol))
brewerMap = brewer2mpl.get_map(colourscheme,category,maxCol,reverse=reverse).colors
# depth,brewerLevel = 0,0
# while depth <= maxRec:
# for x in xrange(0,levels_per_colour):
# depth+=1 ## starts at 1
# depth2RGB[depth] = brewerMap[brewerLevel]
# brewerLevel+=1 ## starts at 0
for depth in xrange(1,maxRec+1):
if depth<maxCol:
depth2RGB[depth] = brewerMap[depth-1]
else:
depth2RGB[depth] = brewerMap[maxCol-1]
## apply it to the regions_merged
for taxa in self.regions_merged:
for idx,x in enumerate(self.regions_merged[taxa]):
self.regions_merged[taxa][idx]['RGB'] = depth2RGB[ x['depth'] ]
def testFilterLinearCoh(n_iterations=5):
cohs = np.array([-0.512, -0.256, -0.128, -0.064, -0.032,
0, 0.032, 0.064, 0.128, 0.256, 0.512])
inputs = []
for coh in cohs:
for j in xrange(n_iterations):
seed = [np.random.randint(999), np.random.randint(999)]
if np.sign(coh) == 1:
direction = 0
else:
direction = 180
inputs.append(
gd.generateDots(seed, 0, 0, 5, 60, direction, np.abs(coh), 5))
pool = multiprocessing.Pool()
out = pool.map(filterDots, inputs)
pool.close()
data = np.array(out)
summed_me = np.sum(data, axis=1)
mean_me = np.mean(summed_me.reshape(-1, n_iterations), axis=1)
fig = plt.figure()
ax = fig.add_subplot(111)
colors = b2mpl.get_map('Set1', 'Qualitative', 3).mpl_colors
ax.add_line(plt.Line2D(cohs, mean_me, color=colors[0]))
ax.set_xlabel('Motion strength (proportion coh)')
ax.set_ylabel('Total motion energy')
ax.set_xlim(-.61, .61)
ax.set_ylim(np.min(mean_me) * 1.2, np.max(mean_me) * 1.2)
sns.despine(offset=5, trim=True)
plt.tight_layout()
return cohs, mean_me
def get_colors(nodes):
n_colors = 8
colors = {}
set2 = brewer2mpl.get_map('Dark2', 'qualitative', n_colors).hex_colors
for i, node in enumerate(nodes):
colors[node] = set2[i % n_colors]
return colors
def plotTimeseries(ts,myvar):
ts_annual = ts.resample("A")
ts_quarterly = ts.resample("Q")
ts_monthly = ts.resample("M")
# Write data to file
mypath="%s_annualaverages.csv"%(myvar)
if os.path.exists(mypath):os.remove(mypath)
ts.to_csv(mypath)
print "Wrote timeseries to file: %s"%(mypath)
red_purple = brewer2mpl.get_map('RdPu', 'Sequential', 9).mpl_colormap
colors = red_purple(np.linspace(0, 1, 12))
fig = plt.figure(figsize=(6,6))
ax = fig.add_subplot(111)
# for mymonth in xrange(12):
#ts[(ts.index.month==mymonth+1)].plot(marker='o', color=colors[mymonth],markersize=5, linewidth=0,alpha=0.8)
#hold(True)
ts_annual.plot(marker='o', color="#FA9D04", linewidth=0,alpha=1.0, markersize=7, label="Annual")
remove_border(top=False, right=False, left=True, bottom=True)
#ts_monthly.plot(style="r", marker='o', linewidth=1,label="Monthly")
# legend(loc='best')
ylabel(r'Light (W m$^{-2})$')
plotfile='figures/timeseries_'+str(myvar)+'.pdf'
plt.savefig(plotfile,dpi=300,bbox_inches="tight",pad_inches=0)
print 'Saved figure file %s\n'%(plotfile)
plt.show()
def struct_to_fig(data):
for a in data:
K = int(data[a]['assn'].shape[0])
bmap = brewer2mpl.get_map('Set3', 'qualitative', K)
colors = bmap.mpl_colors
width=1
xcoord = np.arange(len(data[a]['ids'])*width, step=width)
bars=[]
xcoord = np.arange(len(data[a]['ids'])*width, step=width)
for i, (c, ki) in enumerate(zip(colors,data[a]['assn'])):
if i ==0:
bars.append( plt.bar(left=xcoord, height=ki, width=width, color=c, edgecolor='none') )
else:
bottom = np.sum(data[a]['assn'][:i,:], axis=0)
bars.append( plt.bar(left=xcoord, height=ki, width=width, color=c, bottom=bottom, edgecolor='none') )
plt.title('Structure Plot, K='+str(K))
plt.grid(False)
plt.xlim([0,xcoord[-1]+width])
plt.ylim([0,1])
plt.xticks(xcoord+width/2., data[a]['ids'], rotation=90, fontsize=6)
plt.tick_params(axis='x', which='both', bottom='on', top='off', direction='out')
plt.tick_params(axis='y', which='both', left='off', right='off', labelleft='off')
plt.show()
def esat_comparison_plot(t=_np.linspace(173.15, 373.15, 20), std="Hyland_Wexler", percent=True, log=False):
import pylab
import brewer2mpl
methods = list(esat(0, method="return"))
methods.remove(std)
print len(methods)
pylab.rcParams.update({"axes.color_cycle": brewer2mpl.get_map("Paired", "qualitative", 12).mpl_colors})
y = esat(t, method=std, info=False)
i = 0
style = "-"
for im in methods:
if i > 11:
style = "--"
print im
if percent:
pylab.plot(t, 100.0 * (esat(t, method=im, info=False) - y) / y, lw=2, ls=style)
else:
pylab.plot(t, esat(t, method=im, info=False) - y, lw=2, ls=style)
i += 1
pylab.legend(methods, loc="upper right", fontsize=8)
pylab.xlabel("Temperature [K]")
if percent:
# pylab.semilogy()
pylab.ylabel("Water Vapor Pressure Difference [%]")
else:
pylab.ylabel("Water Vapor Pressure [Pa]")
pylab.title("Comparison of Water Vapor Calculations Ref:" + std)
pylab.xlim(_np.round(t[0]), _np.round(t[-1]))
pylab.grid()
pylab.axvline(x=273.15, color="k")
if log:
pylab.yscale("log")
def data(levels=5,depth=20):
'''return list contour paths calculated from the sum file'''
global file_count,sum_files
sum_files()
paths,matplotlib_colors,legend = get_contour_paths('data/_sumfile.csv',depth)
n=levels
colors = brewer2mpl.get_map('YlOrBr', 'Sequential', n)
result_list = []
hexcolors = colors.hex_colors
m = len(hexcolors)
result_list = []
for i in range(n):
keyval = (float(legend[i])/60.)/file_count
working_dict = {'strokeColor':almostblack,
'fillColor':hexcolors[i],
'fillOpacity':0.4,
'keyValue':'%0.0f min' % keyval,
'polygon':list()}
tmp = list() #all polygons of this level are in the same set
for n,path in enumerate(paths[i]):
p = map(list,path)
tmp.append(p)
if i != 0:
#cut out the hole for the layer before it also
for n,path in enumerate(paths[i-1]):
p = map(list,path)
p.reverse()
tmp.append(p)
working_dict['polygon'].append(tmp)
result_list.append(working_dict)
return result_list
def get_series(values, intervals, color, color_type, reverse=False):
classification_values = []
for v in values:
classification_values.append(float(v['freq']))
classes = classify_values(classification_values, intervals)
#bmap = brewer2mpl.get_map('RdYlGn', 'Diverging', 9, reverse=True)
bmap = brewer2mpl.get_map(color, color_type, intervals+1, reverse=reverse)
colors = bmap.hex_colors
# creating series
series = []
for color in colors:
#print color
series.append({
"color": color,
"data" : []
})
#print classes
for v in values:
freq = v['freq']
for i in range(len(classes)):
if freq <= classes[i]:
series[i]['data'].append([float(v['data'][0]), float(v['data'][1])])
break
return series
def fig_boxplotcomparison(regular, phospho, sheet, column, bounds, figpath):
"""
Figure 2c from Aurora paper
"""
#%%
xvals = []
yvals = []
for i, j in zip([regular[sheet], regular[column], phospho[sheet].values, phospho[column].values], ["peptides", "norm_peptides", "phospho", "norm_phospho"]):
xvals.extend(i)
yvals.extend([j]*len(i))
df = pd.DataFrame([xvals, yvals]).transpose()
df.columns = [column, "peptide type"]
#%%
lower_bound = bounds[0]
upper_bound = bounds[1]
bmap = brewer2mpl.get_map('Paired', 'Qualitative', 6).mpl_colors
f, ax = plt.subplots(1, figsize=(11.69, 8.27))
sns.boxplot(x="peptide type", y=column, data=df, palette=bmap)
ax.set(xticks=[0, 1, 2, 3],
xticklabels=["peptides", "peptides \n(normalized)", "phospho-\npeptides", "phosphopeptides \n(normalized)"])
ax.set(ylim=(-2, 2), ylabel="log2 (fold change)", title=column)
ax.axhline(lower_bound, ls="--", lw=2, color="red", alpha=0.7)
ax.axhline(upper_bound, ls="--", lw=2, color="red", alpha=0.7)
ax.yaxis.set_major_locator(MaxNLocator(4))
sns.despine()
cutils.save_fig(f, figpath+"162_BoxplotNorm_{}".format(column))
def heatmap(file, theta_inc, theta_max):
#env.log("Start ploting heatmap for {} ...".format(file), flush=True)
if os.path.getsize(file) == 0:
hinton('{}.png'.format(file))
return
lods = []
with open(file, 'r') as f:
for line in f.readlines():
theta,lod = line.split()[-2:]
if float(theta) >= theta_max:
continue
lods.append(lod)
if max(lods) == min(lods):
#env.log('Max equals Min for [{}], No real heatmap will be generated.'.format(file))
hinton('{}.png'.format(file))
return
Num=int(round(theta_max/theta_inc))
lods = np.array(map(float,lods)).reshape((-1,Num))
chrID = re.search(r'\.chr([0-9XY]+)\.', file).group(1)
fig, ax = plt.subplots(1)
ax.set_title('Chromosome {}'.format(chrID))
ppl.pcolormesh(fig,ax,lods.transpose(),
xticklabels=[''] * len(lods),
yticklabels=np.round(np.array(range(Num)) * theta_inc,2).tolist(),
cmap=brewer2mpl.get_map('Blues', 'Sequential', 9).mpl_colormap)
fig.savefig('{}.png'.format(file))
def main():
data = generate_series()
print(mpl.rcParams["axes.color_cycle"])
# brewer version
# brewer2mpl.get_map args: set name set type number of colors
bmap = brewer2mpl.get_map("Set1", "qualitative", 9)
# Change the default colors
mpl.rcParams["axes.color_cycle"] = bmap.mpl_colors
plot_data(data)
print(bmap.mpl_colors)
# mpl version
cmap = cm.get_cmap("Set1")
assert isinstance(cmap, LinearSegmentedColormap)
print(type(cmap))
#mpl.rcParams["axes.color_cycle"] =
plot_data(data)
plt.show()
pass
def testFilterSpeedBandwidth(n_iterations=5):
speeds = np.arange(1, 9.5, 0.5)
inputs = []
for speed in speeds:
for j in xrange(n_iterations):
seed = [np.random.randint(999), np.random.randint(999)]
inputs.append(
gd.generateDots(seed, 0, 0, 5, 60, 0, 1, speed))
pool = multiprocessing.Pool()
out = pool.map(filterDots, inputs)
pool.close()
data = np.array(out)
summed_me = np.sum(data, axis=1)
mean_me = np.mean(summed_me.reshape(-1, n_iterations), axis=1)
fig = plt.figure()
ax = fig.add_subplot(111)
colors = b2mpl.get_map('Set1', 'Qualitative', 3).mpl_colors
ax.add_line(plt.Line2D(speeds, mean_me, color=colors[0]))
ax.set_xlabel('Motion speed (degrees per s))')
ax.set_ylabel('Total motion energy')
majorLocator = MultipleLocator(1)
ax.xaxis.set_major_locator(majorLocator)
majorLocator = MultipleLocator(50)
ax.yaxis.set_major_locator(majorLocator)
ax.set_xlim(1, 9)
ax.set_ylim(np.min(mean_me) - 10, np.max(mean_me) + 10)
sns.despine(offset=5, trim=True)
plt.tight_layout()
return speeds, mean_me
def __radd__(self, gg):
gg = deepcopy(gg)
if self.type:
ctype = self.type
else:
ctype = "Sequential"
ctype = _handle_shorthand(ctype)
if self.palette:
palette = self.palette
else:
palette = _number_to_palette(ctype, 1)
if isinstance(palette, int):
palette = _number_to_palette(ctype, palette)
# Try to get colors
try:
color_col = gg.aesthetics.get('color', gg.aesthetics['fill'])
n_colors = max(gg.data[color_col].nunique(),3)
except KeyError :
# If we are neither using 'color' nor 'fill' then assume there is
# only one color used
n_colors = 3
bmap = brewer2mpl.get_map(palette, ctype, n_colors)
gg.manual_color_list = bmap.hex_colors
return gg
def make_pretty_generation_plot(filename, generations, lines_to_plot, x_axis_name, legend_labels, y_errorbar=None, y_logscale = False, vline_x = [], save_figure = True):
if y_errorbar:
assert isinstance(y_errorbar, list), "Please provide a list for error bars"
assert len(y_errorbar) == len(lines_to_plot), "When plotting error bars you must specify an array of error bars for each line that is plotted"
cmap = brewer2mpl.get_map('Set1', 'qualitative', 9)
p = Ppl(cmap, alpha=1)
fig, ax = plt.subplots(1)
ax.set_xlabel('Generation')
ax.set_ylabel(x_axis_name)
if y_logscale:
ax.set_yscale('log')
plt.grid(b=True, which='both', axis='y', color='0.65',linestyle='dashed')
legend_labels = map(lambda s: s + ' (log)', legend_labels)
else: ax.yaxis.get_major_formatter().set_powerlimits((0, 1))
for i, (series, label) in enumerate(zip(lines_to_plot, legend_labels)):
p.plot(ax, generations, series, linewidth=2, label=label, zorder=2)
if y_errorbar:
for i, (series, label) in enumerate(zip(lines_to_plot, legend_labels)):
ax.errorbar(generations, series, yerr=y_errorbar[i], zorder=1, capsize=2, barsabove=True, ecolor = errorbar_color)
if vline_x:
for x in vline_x:
ax.vlines(x = x, ymin = ax.get_ylim()[0], ymax = ax.get_ylim()[1], linewidth = 1, linestyles = 'dashed', alpha = 0.5)
p.legend(ax, loc=0)
if save_figure: fig.savefig(filename)
return fig, ax, filename
def issue_21_basic_scatter_plots(dataset):
"""
Makes scatter plots of fitness
"""
from plotting import make_color_grouped_scatter_plot
folder = make_issue_specific_figure_folder('21_scatter_plots', dataset)
fit_data, par_data, gen, ids = IO.load_pickled_generation_dataframe(dataset_name=dataset)
#colormap = brewer2mpl.get_map('YlOrRd', 'Sequential', 9, reverse=True)
colormap = brewer2mpl.get_map('Spectral', 'Diverging', 9, reverse=True)
print "Making scatter plots of fitness data for dataset %s"%dataset
filename = folder + 'a.png'
make_color_grouped_scatter_plot(data_frame=fit_data, x_name='overshoot', y_name='time_to_reach_new_fundamental', color_by='stdev', filename=filename, colormap = colormap, y_function='log')
filename = folder + 'b.png'
make_color_grouped_scatter_plot(data_frame=fit_data, x_name='overshoot', y_name='stdev', color_by='time_to_reach_new_fundamental', filename=filename, colormap = colormap)
filename = folder + 'c.png'
make_color_grouped_scatter_plot(data_frame=fit_data, x_name='time_to_reach_new_fundamental', y_name='round_stable', color_by='stdev', filename=filename, colormap = colormap)
filename = folder + 'd.png'
make_color_grouped_scatter_plot(data_frame=fit_data, x_name='stdev', y_name='round_stable', color_by='time_to_reach_new_fundamental', filename=filename, colormap = colormap, x_function='log', y_function='log')
filename = folder + 'e.png'
make_color_grouped_scatter_plot(data_frame=fit_data, x_name='stdev', y_name='time_to_reach_new_fundamental', color_by='round_stable', filename=filename, colormap = colormap, x_function='log', y_function='log')
filename = folder + 'f.png'
make_color_grouped_scatter_plot(data_frame=fit_data, x_name='time_to_reach_new_fundamental', y_name='stdev', color_by='round_stable', filename=filename, colormap = colormap)
filename = folder + 'g.png'
make_color_grouped_scatter_plot(data_frame=fit_data, x_name='time_to_reach_new_fundamental', y_name='stdev', color_by='round_stable', filename=filename, colormap = colormap, x_function='log', y_function='log', color_function='log')
def anom_cmap():
"""return a discrete blue-red cmap from colorbrewer"""
ncols = 11
cmap_anom = brewer2mpl.get_map('RdBu', 'diverging', ncols,
reverse=True).mpl_colormap
cmap_anom = discrete_cmap(ncols, cmap_anom)
return cmap_anom
def plot_figures(fignum='2a', direction='-', show=False,
T=45., R=0.1, gamma=1.):
import brewer2mpl
cmap = brewer2mpl.get_map('Set1', 'qualitative', 9)
colors = cmap.mpl_colors
params = {"T": T,
"R": R,
"gamma": gamma,
"loop_type": "Outside_circle"}
if fignum == '2a':
settings = {"init_state": 'a',
"init_phase": 0,
"loop_direction": '+'}
elif fignum == '2b':
settings = {"init_state": 'b',
"init_phase": 0,
"loop_direction": '-'}
elif fignum == '2c':
settings = {"init_state": 'b',
"init_phase": np.pi,
"loop_direction": '-'}
params.update(settings)
OM = OptoMech(**params)
t, cp, cm = OM.solve_ODE()
Psi = OM.Psi
# f, (ax1, ax2) = plt.subplots(ncols=2)
f, ax1 = plt.subplots()
ax1.semilogy(t, np.abs(Psi[:, 0])**2, ls="-", ms="o",
color=colors[0], label=r"$|\alpha_+(t)|^2$")
ax1.semilogy(t, np.abs(Psi[:, 1])**2, ls="-", ms="o",
color=colors[1], label=r"$|\alpha_-(t)|^2$")
ax1.semilogy(t, np.abs(cp)**2, ls="--",
color=colors[2], label=r"$|c_+(t)|^2$")
ax1.semilogy(t, np.abs(cm)**2, ls="--",
color=colors[3], label=r"$|c_-(t)|^2$")
ax1.legend(loc="lower right")
ax1.set_xlabel(r"$t$")
m = [(abs(x)**2).max() for x in Psi, cp, cm]
ax1.set_ylim(1e-3, max(m))
omega, g = OM.get_cycle_parameters(t)
# np.savetxt("parameters_{}.dat".format(fignum), zip(g, omega))
ax2 = plt.axes([0.2, 0.65, .2, .2])
ax2.plot(gamma/2, 0, "ko")
ax2.plot(g, omega, ls="-", color=colors[0])
ax2.set_xlim(gamma/4, 3/4.*gamma)
ax2.set_ylim(-gamma/4., gamma/4.)
if show:
plt.show()
else:
plt.savefig("{}.png".format(fignum))
def plot_integrated_colors(filenames, labels='Z'):
if type(filenames) is str:
filenames = [filenames]
ax = None
cols = ['k']
else:
fig, ax = plt.subplots()
cols = brewer2mpl.get_map('Spectral', 'Diverging',
len(filenames)).mpl_colors
if labels == 'Z':
fmt = '$Z=%.4f$'
labels = [fmt % float(l.replace('.dat', '').split('Z')[1])
for l in filenames]
else:
print 'need to fix labels'
labels = [''] * len(filenames)
for i, filename in enumerate(filenames):
data = rsp.fileIO.readfile(filename)
ycol = 'V-K'
xcol = 'Age'
ax = rg.color_color(data, xcol, ycol, xscale='log', ax=ax,
plt_kw={'lw': 2, 'color': cols[i],
'label': labels[i]})
plot_cluster_data(ax)
ax.legend(frameon=False, loc=0, numpoints=1)
ax.set_xlabel(r'${\rm %s}$' % xcol, fontsize=20)
ax.set_ylabel(r'${\rm %s}$' % ycol, fontsize=20)
plt.tick_params(labelsize=16)
return ax
def get_colors(non_null):
if 'brewer2mpl' in globals():
num_color = max(min(len(non_null), 11), 3)
all_colors = brewer2mpl.get_map('Spectral', 'Diverging', num_color).hex_colors
else:
all_colors = None
return all_colors
def paint_clusters(G1, G2, beacons_G1, beacons_G2, kmeans_labels1, kmeans_labels2):
accent_colors = brewer2mpl.get_map('Accent', 'qualitative',8).mpl_colors
plt.subplot(121)
plt.axis('off')
plt.title('$G_1$')
(labels, node_colors, node_sizes) = visualize_beacons(G1, beacons_G1)
clustered_colors = cluster_colors(G1, beacons_G1, kmeans_labels1, accent_colors)
pos = nx.spring_layout(G1, weight=None)
nx.draw_networkx_nodes(G1, pos, node_color=clustered_colors, node_size=node_sizes, font_size=18)
nx.draw_networkx_labels(G1, pos, font_size=17, labels=labels, font_color = '#262626')
nx.draw_networkx_edges(G1, pos, width=2, alpha=0.3)
plt.subplot(122)
plt.axis('off')
plt.title('$G_2$')
(labels, node_colors, node_sizes) = visualize_beacons(G2, beacons_G2)
clustered_colors = cluster_colors(G2, beacons_G2, kmeans_labels2, accent_colors)
pos2_init= {key:value for (key, value) in zip(beacons_G2,[pos[beacon] for beacon in beacons_G1])}
pos2 = nx.spring_layout(G2, weight=None, pos = pos2_init)
nx.draw_networkx_nodes(G2, pos=pos2, node_color=clustered_colors, node_size=node_sizes, font_size=18)
nx.draw_networkx_labels(G2, pos=pos2, font_size=17, labels=labels, font_color = '#262626')
nx.draw_networkx_edges(G2, pos=pos2, width=2, alpha=0.3)
import cPickle
import matplotlib.pyplot as plt
import numpy as np
import sys
import brewer2mpl
bmap = brewer2mpl.get_map('Dark2', 'qualitative', 6)
colors = bmap.mpl_colors
colors_config = {
"hand": colors[1],
"stick": colors[2],
"object": colors[4],
}
bmap2 = brewer2mpl.get_map('Dark2', 'qualitative', 4)
colors2 = bmap2.mpl_colors
plt.switch_backend('Agg')
sw = 1
def runningMeanFast(x, sw):
return np.convolve(x, np.ones((sw, )) / sw, mode="valid")
def n_behavior_change(x, smoothing=10, th=10):
x = runningMeanFast(x, smoothing)
l = np.linspace(0, len(x) - 1, 1 + int((len(x) - 1) / smoothing))
l = np.array(l, dtype=int)
plt.yticks(range(1, 24))
yticks = range(1, 23)
yticks.append('X')
ax.set_yticklabels(yticks)
else:
plt.yticks(range(1, 23))
#define colors
def hex_to_rgb(value):
value = value.lstrip('#')
lv = len(value)
return tuple(int(value[i:i + lv // 3], 16) for i in range(0, lv, lv // 3))
bmap = brewer2mpl.get_map('Set1', 'qualitative', 4)
if args.colors is not None:
colors = []
color_list = args.colors.split(',')
[colors.append(x) for x in color_list]
else:
colors = bmap.mpl_colors
colors.append((0, 0, 0))
#define centromeres
centro = open(args.centromeres)
centromeres = {}
for line in centro:
line = line.strip().split()
if chrX and line[0] == 'X':
line[0] = '23'
import os
import sys
import cPickle
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import brewer2mpl
bmap = brewer2mpl.get_map('Dark2', 'qualitative', 8)
#colors = bmap.mpl_colors
colors = {
"motor_babbling": bmap.mpl_colors[0],
"Hand": bmap.mpl_colors[1],
"Joystick_L": bmap.mpl_colors[2],
"Joystick_R": bmap.mpl_colors[3],
"Ergo": bmap.mpl_colors[4],
"Ball": bmap.mpl_colors[5],
"Light": bmap.mpl_colors[6],
"Sound": bmap.mpl_colors[7]
}
simu = False
if simu:
# From SIMU
path = "/home/sforesti/catkin_ws/src/nips2017/logs/"
experiment_name = "experiment"
configs = dict(AMB=9) #, RMB=3, RmB=3, FC=1, OS=3)
n = 10000
def plot():
cs = palettable.colorbrewer.qualitative.Set1_3.mpl_colors
# Load parameter file
with open('/home/lc585/qsosed/input.yml', 'r') as f:
parfile = yaml.load(f)
ebvmin = -0.075
ebvmax = 0.075
zmin = 1
zmax = 3.0
# Load stuff
fittingobj = load(parfile)
wavlen = fittingobj.get_wavlen()
# Load data
df = get_data()
df = df[(df.z_HW >= zmin) & (df.z_HW <= zmax)]
colmin, colmax = [], []
zs = np.arange(zmin, zmax + 0.025, 0.025)
mycm = cm.get_cmap('Blues_r')
mycm.set_under('w')
cset = brewer2mpl.get_map('Blues', 'sequential', 9).mpl_colors
mycm = truncate_colormap(mycm, 0.0, 0.8)
fig = plt.figure(figsize=(figsize(1, vscale=0.8)))
ax = fig.add_subplot(1, 1, 1)
plt.tight_layout()
#histogram definition
xyrange = [[0., 3.0], [0, 3]] # data range
bins = [100, 60] # number of bins
thresh = 7 #density threshold
#data definition
xdat, ydat = df.z_HW, df.iVEGA - df.KVEGA
# histogram the data
hh, locx, locy = histogram2d(xdat, ydat, range=xyrange, bins=bins)
posx = np.digitize(xdat, locx)
posy = np.digitize(ydat, locy)
#select points within the histogram
ind = (posx > 0) & (posx <= bins[0]) & (posy > 0) & (posy <= bins[1])
hhsub = hh[posx[ind] - 1,
posy[ind] - 1] # values of the histogram where the points are
xdat1 = xdat[ind][hhsub < thresh] # low density points
ydat1 = ydat[ind][hhsub < thresh]
hh[hh < thresh] = np.nan # fill the areas with low density by NaNs
im = ax.imshow(np.flipud(hh.T),
cmap=mycm,
extent=np.array(xyrange).flatten(),
interpolation='none',
aspect='auto',
vmin=thresh,
vmax=45)
clb = fig.colorbar(im)
clb.set_label('Number of Objects')
ax.scatter(xdat1, ydat1, color=cset[-1], s=2)
plt.ylim(0, 3.5)
col = []
for z in zs:
parfile['ext']['EBV'] = ebvmax
magtmp, wavlentmp, fluxtmp = model(redshift=z, parfile=parfile)
col.append(magtmp[3] - magtmp[8])
plt.plot(zs, col, label='E(B-V) = 0.075', color=cs[0], linewidth=1.0)
upper_bound_1 = col[np.argmin(np.abs(zs - 1.0)):np.argmin(np.abs(zs -
1.5))]
upper_bound_2 = col[np.argmin(np.abs(zs - 2.0)):np.argmin(np.abs(zs -
2.7))]
col = []
for z in zs:
parfile['ext']['EBV'] = ebvmin
magtmp, wavlentmp, fluxtmp = model(redshift=z, parfile=parfile)
col.append(magtmp[3] - magtmp[8])
plt.plot(zs, col, label='E(B-V) = -0.075', color=cs[0], linewidth=1.0)
lower_bound_1 = col[np.argmin(np.abs(zs - 1.0)):np.argmin(np.abs(zs -
1.5))]
lower_bound_2 = col[np.argmin(np.abs(zs - 2.0)):np.argmin(np.abs(zs -
2.7))]
plt.fill_between(zs[np.argmin(np.abs(zs - 2.0)):np.argmin(np.abs(zs -
2.7))],
lower_bound_2,
upper_bound_2,
facecolor='None',
edgecolor=cs[0],
linewidth=3.0)
col = []
for z in zs:
parfile['ext']['EBV'] = 0.0
magtmp, wavlentmp, fluxtmp = model(redshift=z, parfile=parfile)
col.append(magtmp[3] - magtmp[8])
plt.plot(zs, col, label='E(B-V) = 0.0', color=cs[0], linewidth=1.0)
plt.xlim(0.75, 3.5)
plt.text(3.33821,
2.5,
'E(B-V)=',
horizontalalignment='right',
verticalalignment='center',
color='black')
plt.text(3.09608,
2.2,
'0.075',
horizontalalignment='left',
verticalalignment='center',
color='black')
plt.text(3.09608,
1.2,
'-0.075',
horizontalalignment='left',
verticalalignment='center',
color='black')
plt.text(3.09608,
1.7,
'0.0',
horizontalalignment='left',
verticalalignment='center',
color='black')
plt.xlabel(r'Redshift $z$')
plt.ylabel(r'$i$-$K$')
plt.tight_layout()
fig.savefig('/home/lc585/thesis/figures/chapter05/ik_versus_z_low_ext.pdf')
plt.show()
return None
import glob
from pylab import *
import scipy.stats
import brewer2mpl
bmap = brewer2mpl.get_map('Set2', 'qualitative', 7)
colors = bmap.mpl_colors
params = {
'axes.labelsize': 8,
'font.size': 8,
'legend.fontsize': 10,
'xtick.labelsize': 10,
'ytick.labelsize': 10,
'text.usetex': False,
'figure.figsize': [2.5, 4.5]
}
rcParams.update(params)
def load(dir):
f_list = glob.glob(dir + '/*/*/bestfit.dat')
num_lines = sum(1 for line in open(f_list[0]))
i = 0
data = np.zeros((len(f_list), num_lines))
for f in f_list:
data[i, :] = np.loadtxt(f)[:, 1]
i += 1
return data
mpl.rcParams["savefig.transparent"] = True
mpl.rcParams["savefig.bbox"] = "tight"
mpl.rcParams["font.size"] = 10.0
mpl.rcParams["font.sans-serif"] = "Arial"
mpl.rcParams["savefig.format"] = "pdf"
import numpy as np
import pandas as pd
from joblib import Parallel, delayed
from sklearn.cluster import DBSCAN
from sklearn import metrics
from sklearn.datasets.samples_generator import make_blobs
from sklearn.preprocessing import StandardScaler
import pylab
import brewer2mpl
#import seaborn as sns
colors = brewer2mpl.get_map('Set3', 'qualitative', 8).mpl_colors
#own library
from cDBSCAN import cDBSCAN
from kDBSCAN import kDBSCAN
def getDataD(r=1, samples=1000, std1=0.2, centercounts=10, rmax=15):
centers = []
for i in xrange(centercounts):
x, y = np.random.randint(0 - rmax,
rmax), np.random.randint(0 - rmax, rmax)
centers.append([x, y])
#data as true clusters
X, xlabels = make_blobs(n_samples=samples,
centers=centers,
h = plt.gca()
h.set_xlabel('epochs')
plt.legend(loc='best')
fig.savefig('{}/{}_{}.png'.format(folder_today, plotx, meas),
bbox_inches='tight')
# %% --- Prepapare visual output as in [1] ---
# set line width and style
lwidth = 2
lwidth_help = 2
lstyle = '-'
lstyle_help = '--'
# set colors using colorbrewer
bmap = brewer2mpl.get_map('Paired', 'Qualitative', 5)
colors = bmap.mpl_colors
# set latex options
matplotlib.rc('text', usetex=True)
matplotlib.rcParams['text.latex.preamble'] = [r"\usepackage{amsmath}"]
# set font
fsize = 15
font = {'family': 'serif', 'size': fsize}
matplotlib.rc('font', **font)
matplotlib.rc('axes', labelsize=fsize) # fontsize of x and y labels
matplotlib.rc('xtick', labelsize=fsize) # fontsize of xtick labels
matplotlib.rc('ytick', labelsize=fsize) # fontsize of ytick labels
matplotlib.rc('legend', fontsize=fsize) # legend fontsize
import time
import numpy as np
np.random.seed(1111)
np.seterr(over="raise")
import cPickle, os
from hips.plotting.layout import create_figure
import matplotlib.pyplot as plt
import brewer2mpl
colors = brewer2mpl.get_map("Set1", "Qualitative", 9).mpl_colors
# goodcolors = np.array([0,1,2,4,6,7,8])
# colors = np.array(colors)[goodcolors]
from pybasicbayes.util.general import ibincount
from pybasicbayes.util.text import progprint_xrange
import pyhawkes.models
reload(pyhawkes.models)
# Set globals
K = 10
B = 3
dt = 1
dt_max = 10.
T = 100.
network_hypers = {
'C': 1,
'kappa': 1.,
from naive_bayes import BernoulliNB, GaussianNB, GaussianNB2, MultinomialNB, PoissonNB, MixNB, MixNB2
from sklearn.linear_model import LogisticRegression, LinearRegression
from sklearn.neighbors import KNeighborsClassifier
from sklearn.svm import SVC
from sklearn.ensemble import RandomForestClassifier
from matplotlib.mlab import rec_drop_fields
from matplotlib import cm
# import itertools
from inspect import getargspec
from sklearn.grid_search import RandomizedSearchCV
import os
import re
from sklearn import preprocessing
import brewer2mpl
from scipy.stats import itemfreq
paired = brewer2mpl.get_map('Paired', 'qualitative', 10).mpl_colors
# Testing Pipeline:
def testAlgo(clf,
X,
y,
clfName,
opt=False,
param_dict=None,
opt_metric='roc_auc',
n_iter=50,
folds=10,
times=10):
'''An algorithm that output the perdicted y and real y'''
y_true = []
t_opts = []
t_stars = []
phi_stars = []
#recover the parts
for result in results:
delta_phi_fs.append(result[0])
t_opts.append(result[1])
t_stars.append(result[2])
phi_stars.append(result[3])
delta_phi_fs = np.vstack(delta_phi_fs).T
t_opts = np.vstack(t_opts).T
phi0s_plt = np.vstack([phi0s] * 200)
np.save('Data/t_opts.npy', t_opts)
palatte = colorbrewer.get_map('Dark2', 'Qualitative', 3)
colors = palatte.mpl_colors
plo.PlotOptions(uselatex=True, ticks='in')
plt.figure(figsize=(3.5, 2.85))
gs = gridspec.GridSpec(3,
2,
height_ratios=(1, 0.5, 0.5),
width_ratios=(1.5, 1))
ax = plt.subplot(gs[0, 0])
ts = np.arange(0, 23.7, 0.01)
ax.plot(times * np.pi / 11.85, prc, color='k')
ax.set_ylabel('ipPRC$(\phi)$')
ax.set_xlabel('Phase $\phi$')
plo.format_2pi_axis(ax)
p.add_argument('--output',
'-o',
help='Output to file. Otherwise show.',
nargs='?')
args = p.parse_args()
if args.style:
try:
from mpltools import style
style.use(args.style)
except ImportError:
print("Need mpltools for setting styles (pip install mpltools)")
try:
import brewer2mpl
all_colors = brewer2mpl.get_map('Paired', 'Qualitative', 12).hex_colors
except ImportError:
print("Install brewer2mpl for better colors (pip install brewer2mpl)")
all_colors = (
'green',
'orange',
'red',
'blue',
'black',
'olive',
'purple',
'#6960EC',
'#F0FFFF',
'#728C00',
'#827B60',
'#F87217',
def perfcheck(nstart=0,nend=100,noisemax=100):
bmap = brewer2mpl.get_map('Set2','qualitative', 7)
colors = bmap.mpl_colors
params = {
'axes.labelsize': 15,
'font.size': 12,
'legend.fontsize': 15,
'xtick.labelsize': 20,
'ytick.labelsize': 20,
'text.usetex': True ,
'figure.figsize': [8, 6], # instead of 4.5, 4.5
'font.weight': 'bold',
'axes.labelweight': 'bold',
'ps.useafm' : True,
'pdf.use14corefonts':True,
'pdf.fonttype': 42,
'ps.fonttype': 42
}
mpl.rcParams.update(params)
# y=np.array(np.loadtxt('perfcheck.txt'))
# perf=np.mean(y,axis=1)
# cstd=np.std(y,axis=1)
# step=noisemax/int(perf.shape[0]-1)
# sind=int(nstart/step)
# eind=int(nend/step)+1
# plt.figure(figsize=(12,9))
# f5,=plt.plot(np.arange(sind,(eind-1)*step+1,step),perf[sind:eind],color=colors[0], linewidth=4)
# plt.fill_between(np.arange(sind,(eind-1)*step+1,step),(perf[sind:eind]-cstd[sind:eind]),(perf[sind:eind]+cstd[sind:eind]), alpha=0.3, color=colors[0])
#D2C
i, x, y = np.loadtxt('swimmer6_testing_data.txt', dtype=np.float64, delimiter=',\t', unpack=True, usecols=(0,1,2))
i = 100*i
plt.figure(1)
plt.plot(i,x,color=colors[1], linewidth=4)
#plt.plot(i, (x+y), alpha=0.3, color='orange')
#plt.plot(i, (x-y), alpha=0.3, color='orange')
plt.fill_between(i, (x+y), (x-y), alpha=0.2, color=colors[1])
#plt.xlabel(" Percent of max. control (Std dev of perturbed noise)", fontsize=16)
#plt.ylabel("Terminal state MSE (Avergaed over {} samples)".format(n_samples), fontsize=16)
plt.grid(axis='y', color='.910', linestyle='-', linewidth=1.5)
plt.grid(axis='x', color='.910', linestyle='-', linewidth=1.5)
#DDPG
DDPG_testing_data_location = "/media/karthikeya/Elements/DDPG_D2C/results/Swimmer6/exp_4/"
i, x, y, z, a = np.loadtxt(DDPG_testing_data_location + "data.txt", dtype=np.float64, delimiter=',\t', unpack=True, usecols=(0,1,2,3,4))
i = 100*i
plt.plot(i,x,color=colors[2], linewidth=4)
#plt.plot(i, (x+y), alpha=0.3, color='orange')
#plt.plot(i, (x-y), alpha=0.3, color='orange')
plt.fill_between(i, (x+y), (x-y), alpha=0.2, color=colors[2])
#plt.xlabel(" Percent of max. control (Std dev of perturbed noise)", fontsize=16)
#plt.ylabel("Terminal state MSE (Avergaed over {} samples)".format(n_samples), fontsize=16)
plt.grid(axis='y', color='.910', linestyle='-', linewidth=1.5)
plt.grid(axis='x', color='.910', linestyle='-', linewidth=1.5)
#plt.tight_layout()
plt.legend(['D2C','DDPG'], loc='upper left', fontsize='xx-large')
plt.xlabel('Std dev of perturbed noise (Percent of max. control)', fontweight='bold',fontsize=22)
plt.ylabel('L2-norm of terminal state error', fontweight='bold', fontsize=25)
plt.grid(True)
plt.show()
print('averaged by {value1} rollouts'.format(value1=y.shape[1]))
lick3311 = lick[:, idx3311]
lickhalo = lick[:, idxhalo]
errs1_3311 = lickerr[:, idx3311]
errs_halo = lickerr[:, idxhalo]
#########################################################################
# First figure, simple indices
app = "_pa" if restrict_pa else ""
mkfig1 = True
gray = "0.75"
##########################################################################
lims, ranges = get_model_lims(
os.path.join(tables_dir, "models_thomas_2010_metal_extrapolated.dat"))
idx = np.array([12, 13, 16, 17, 18, 19, 20])
lims = lims[idx]
# Setting the colormap properties for the scatter plots
cmap = brewer2mpl.get_map('Blues', 'sequential', 9).mpl_colormap
cmap = nc.cmap_discretize(cmap, 3)
color = cm.get_cmap(cmap)
norm = Normalize(vmin=0, vmax=45)
if mkfig1:
plt.figure(1, figsize=(6, 14))
gs = gridspec.GridSpec(7, 1)
gs.update(left=0.15,
right=0.95,
bottom=0.1,
top=0.94,
wspace=0.1,
hspace=0.09)
tex = []
for j, ll in enumerate(lick):
# print indices[j], ranges[j], ssp.fn(9.,0.12,.4)[ii[j]]
def get_colors():
"""
Get colorbrewer colors, which are nicer
"""
bmap = brewer2mpl.get_map('Accent', 'qualitative', 6)
return bmap.mpl_colors
if 'data_contacts' not in globals():
sns_contacts,(data_contacts,calc_contacts) = work.sns(),plotload('contacts',work)
sns_hbonds,(data_hbonds,calc_hbonds) = work.sns(),plotload('hydrogen_bonding',work)
if sns_hbonds!=sns_contacts:
raise Exception('collections for hydrogen_bonding and contacts are not equal')
else: sns = sns_hbonds
#---set the cutoff in the yaml file to make this plot because we pull from multiple upstream sources
cutoff = calc_contacts['calcs']['specs']['cutoff']
#---specify the upstream data and naming scheme
rowspec = ['subject_resname','subject_resid','subject_atom',
'target_resname','target_resid','target_atom']
#---common colors
import brewer2mpl
colors = {'DOPC':'blue','DOPS':'red','POP2':'magenta','PI2P':'magenta',
'all lipids':'black','DOPE':'blue'}
colors.update(**dict([(sn,brewer2mpl.get_map('Set1','qualitative',9).mpl_colors[sns.index(sn)])
for sn in sns]))
lipid_label = lambda x: dict([(i,'$$\mathrm{{PIP}_{2}}$$')
for i in work.vars.get('selectors',{}).get('resnames_PIP2',{})]).get(x,x)
sn_title = lambda sn: '%s%s'%(work.meta[sn].get('label',re.sub('_','-',sn)),
'' if not work.meta[sn].get('cholesterol',False) else '\n(cholesterol)')
#---post-post processing
if 'postdat' not in globals():
salt_bridge_filter()
hydrogen_bond_compactor()
postdat = make_postdat()
#---block: main plotting loop (see settings in the header)
for bond in bond_mappings:
for specname,spec in specs.items():
this_plotspec = dict(plotspec)
# Built-in modules #
import os, time, getpass
# Internal modules #
from common import split_thousands
from autopaths import FilePath
# Third party modules #
import matplotlib, brewer2mpl
from matplotlib import pyplot
# Constants #
cool_colors = brewer2mpl.get_map('Set1', 'qualitative', 8).mpl_colors
cool_colors.reverse()
cool_colors += brewer2mpl.get_map('Set2', 'qualitative', 8).mpl_colors
cool_colors += brewer2mpl.get_map('Set3', 'qualitative', 8).mpl_colors
cool_colors += brewer2mpl.get_map('Pastel1', 'qualitative', 8).mpl_colors
cool_colors += brewer2mpl.get_map('Pastel2', 'qualitative', 8).mpl_colors
cool_colors += brewer2mpl.get_map('Greys', 'sequential', 8).mpl_colors
################################################################################
class Graph(FilePath):
width = 12.0
height = 7.0
bottom = 0.14
top = 0.93
left = 0.06
right = 0.98
formats = ('pdf', )
def plot():
tab = Table.read('/data/lc585/QSOSED/Results/141203/sample1/out_add.fits')
tab = tab[tab['BBT_STDERR'] < 200.0]
tab = tab[tab['BBFLXNRM_STDERR'] < 0.05]
tab = tab[tab['CHI2_RED'] < 3.0]
tab = tab[tab['LUM_IR'] > 40.0] # just gets rid of one annoying point
mycm = cm.get_cmap('YlOrRd_r')
mycm.set_under('w')
cset = brewer2mpl.get_map('YlOrRd', 'sequential', 9).mpl_colors
mycm = truncate_colormap(mycm, 0.0, 0.8)
set_plot_properties() # change style
fig = plt.figure(figsize=figsize(1.5, 0.7))
gs = gridspec.GridSpec(9, 13)
ax1 = fig.add_subplot(gs[1:5, 0:4])
ax2 = fig.add_subplot(gs[5:, 0:4])
ax3 = fig.add_subplot(gs[0, 0:4])
ax4 = fig.add_subplot(gs[1:5, 4:8])
ax5 = fig.add_subplot(gs[5:, 4:8])
ax6 = fig.add_subplot(gs[0, 4:8])
ax7 = fig.add_subplot(gs[1:5, 8:12])
ax8 = fig.add_subplot(gs[5:, 8:12])
ax9 = fig.add_subplot(gs[0, 8:12])
ax10 = fig.add_subplot(gs[1:5, 12])
ax11 = fig.add_subplot(gs[5:, 12])
fig.subplots_adjust(wspace=0.0)
fig.subplots_adjust(hspace=0.0)
#histogram definition
xyrange = [[44, 48], [0, 2]] # data range
bins = [80, 40] # number of bins
thresh = 4 #density threshold
#data definition
xdat, ydat = tab['LUM_UV'], tab['RATIO_IR_UV']
# histogram the data
hh, locx, locy = histogram2d(xdat, ydat, range=xyrange, bins=bins)
posx = np.digitize(xdat, locx)
posy = np.digitize(ydat, locy)
#select points within the histogram
ind = (posx > 0) & (posx <= bins[0]) & (posy > 0) & (posy <= bins[1])
hhsub = hh[posx[ind] - 1,
posy[ind] - 1] # values of the histogram where the points are
xdat1 = xdat[ind][hhsub < thresh] # low density points
ydat1 = ydat[ind][hhsub < thresh]
hh[hh < thresh] = np.nan # fill the areas with low density by NaNs
im = ax1.imshow(np.flipud(hh.T),
cmap=mycm,
extent=np.array(xyrange).flatten(),
interpolation='none',
aspect='auto',
vmin=thresh,
vmax=45)
ax1.scatter(xdat1, ydat1, color=cset[-1])
ax1.set_ylabel(r'$R_{{\rm NIR}/{\rm UV}}$', fontsize=14)
ax1.set_ylim(0., 2.)
ax1.set_xlim(45.25, 47)
ax3.hist(tab['LUM_UV'], color=cset[-1], bins=20)
ax3.set_xlim(ax1.get_xlim())
ax3.set_axis_off()
#histogram definition
xyrange = [[44, 48], [200, 2000]] # data range
bins = [90, 35] # number of bins
thresh = 4 #density threshold
#data definition
xdat, ydat = tab['LUM_UV'], tab['BBT']
# histogram the data
hh, locx, locy = histogram2d(xdat, ydat, range=xyrange, bins=bins)
posx = np.digitize(xdat, locx)
posy = np.digitize(ydat, locy)
#select points within the histogram
ind = (posx > 0) & (posx <= bins[0]) & (posy > 0) & (posy <= bins[1])
hhsub = hh[posx[ind] - 1,
posy[ind] - 1] # values of the histogram where the points are
xdat1 = xdat[ind][hhsub < thresh] # low density points
ydat1 = ydat[ind][hhsub < thresh]
hh[hh < thresh] = np.nan # fill the areas with low density by NaNs
im = ax2.imshow(np.flipud(hh.T),
cmap=mycm,
extent=np.array(xyrange).flatten(),
interpolation='none',
aspect='auto',
vmin=thresh,
vmax=45)
ax2.scatter(xdat1, ydat1, color=cset[-1])
ax2.set_xlabel(r'Log$_{10} (L_{\rm UV} ({\rm erg/s}))$', fontsize=14)
ax2.set_ylabel(r'$T_{\mathrm{BB}}$', fontsize=14)
ax2.set_xlim(ax1.get_xlim())
ax2.set_ylim(500, 2100)
plt.tick_params(axis='both', which='major', labelsize=10)
ax1.get_xaxis().set_ticks([])
ax2.get_yaxis().set_ticks([600, 800, 1000, 1200, 1400, 1600, 1800, 2000])
ax2.get_xaxis().set_ticks([45.4, 45.6, 45.8, 46.0, 46.2, 46.4, 46.6, 46.8])
#############################################################################
#histogram definition
xyrange = [[7.5, 10.5], [0, 2]] # data range
bins = [30, 30] # number of bins
thresh = 4 #density threshold
#data definition
xdat, ydat = tab['LOGBH'], tab['RATIO_IR_UV']
# histogram the data
hh, locx, locy = histogram2d(xdat, ydat, range=xyrange, bins=bins)
posx = np.digitize(xdat, locx)
posy = np.digitize(ydat, locy)
#select points within the histogram
ind = (posx > 0) & (posx <= bins[0]) & (posy > 0) & (posy <= bins[1])
hhsub = hh[posx[ind] - 1,
posy[ind] - 1] # values of the histogram where the points are
xdat1 = xdat[ind][hhsub < thresh] # low density points
ydat1 = ydat[ind][hhsub < thresh]
hh[hh < thresh] = np.nan # fill the areas with low density by NaNs
im = ax4.imshow(np.flipud(hh.T),
cmap=mycm,
extent=np.array(xyrange).flatten(),
interpolation='none',
aspect='auto',
vmin=thresh,
vmax=45)
ax4.scatter(xdat1, ydat1, color=cset[-1])
ax4.set_ylim(ax1.get_ylim())
ax4.set_xlim(7.5, 10.5)
ax4.set_xticklabels([])
ax4.set_yticklabels([])
ax6.hist(tab[tab['LOGBH'] > 6]['LOGBH'], color=cset[-1], bins=20)
ax6.set_xlim(ax4.get_xlim())
ax6.set_axis_off()
#histogram definition
xyrange = [[7.5, 10.5], [500, 2000]] # data range
bins = [50, 30] # number of bins
thresh = 4 #density threshold
#data definition
xdat, ydat = tab['LOGBH'], tab['BBT']
# histogram the data
hh, locx, locy = histogram2d(xdat, ydat, range=xyrange, bins=bins)
posx = np.digitize(xdat, locx)
posy = np.digitize(ydat, locy)
#select points within the histogram
ind = (posx > 0) & (posx <= bins[0]) & (posy > 0) & (posy <= bins[1])
hhsub = hh[posx[ind] - 1,
posy[ind] - 1] # values of the histogram where the points are
xdat1 = xdat[ind][hhsub < thresh] # low density points
ydat1 = ydat[ind][hhsub < thresh]
hh[hh < thresh] = np.nan # fill the areas with low density by NaNs
im = ax5.imshow(np.flipud(hh.T),
cmap=mycm,
extent=np.array(xyrange).flatten(),
interpolation='none',
aspect='auto',
vmin=thresh,
vmax=45)
ax5.scatter(xdat1, ydat1, color=cset[-1])
ax5.set_xlabel(r'Log$_{10}$ (Black Hole Mass $M_{\rm BH}$)')
ax5.set_yticklabels([])
# ax5.hist(tab['BBT'], orientation='horizontal',color=cset[-1],bins=20)
# ax5.set_ylim(ax2.get_ylim())
# ax5.set_axis_off()
plt.tick_params(axis='both', which='major')
ax4.set_ylim(ax1.get_ylim())
ax5.set_ylim(ax2.get_ylim())
ax5.set_xlim(ax4.get_xlim())
ax6.hist(tab['LOGBH'][tab['LOGBH'] > 6], color=cset[-1], bins=20)
ax6.set_xlim(ax5.get_xlim())
ax6.set_axis_off()
######################################################################################
#histogram definition
xyrange = [[-2, 0.5], [0, 2]] # data range
bins = [30, 30] # number of bins
thresh = 4 #density threshold
#data definition
xdat, ydat = tab['LOGEDD_RATIO'], tab['RATIO_IR_UV']
# histogram the data
hh, locx, locy = histogram2d(xdat, ydat, range=xyrange, bins=bins)
posx = np.digitize(xdat, locx)
posy = np.digitize(ydat, locy)
#select points within the histogram
ind = (posx > 0) & (posx <= bins[0]) & (posy > 0) & (posy <= bins[1])
hhsub = hh[posx[ind] - 1,
posy[ind] - 1] # values of the histogram where the points are
xdat1 = xdat[ind][hhsub < thresh] # low density points
ydat1 = ydat[ind][hhsub < thresh]
hh[hh < thresh] = np.nan # fill the areas with low density by NaNs
im = ax7.imshow(np.flipud(hh.T),
cmap=mycm,
extent=np.array(xyrange).flatten(),
interpolation='none',
aspect='auto',
vmin=thresh,
vmax=45)
ax7.scatter(xdat1, ydat1, color=cset[-1])
ax7.set_ylim(ax1.get_ylim())
ax7.set_xlim(-2, 0.5)
ax9.hist(tab[tab['LOGEDD_RATIO'] > -2.5]['LOGEDD_RATIO'],
color=cset[-1],
bins=20)
ax9.set_xlim(ax7.get_xlim())
ax9.set_axis_off()
#histogram definition
xyrange = [[-2, 0.5], [500, 2000]] # data range
bins = [50, 30] # number of bins
thresh = 4 #density threshold
#data definition
xdat, ydat = tab['LOGEDD_RATIO'], tab['BBT']
# histogram the data
hh, locx, locy = histogram2d(xdat, ydat, range=xyrange, bins=bins)
posx = np.digitize(xdat, locx)
posy = np.digitize(ydat, locy)
#select points within the histogram
ind = (posx > 0) & (posx <= bins[0]) & (posy > 0) & (posy <= bins[1])
hhsub = hh[posx[ind] - 1,
posy[ind] - 1] # values of the histogram where the points are
xdat1 = xdat[ind][hhsub < thresh] # low density points
ydat1 = ydat[ind][hhsub < thresh]
hh[hh < thresh] = np.nan # fill the areas with low density by NaNs
im = ax8.imshow(np.flipud(hh.T),
cmap=mycm,
extent=np.array(xyrange).flatten(),
interpolation='none',
aspect='auto',
vmin=thresh,
vmax=45)
ax8.scatter(xdat1, ydat1, color=cset[-1])
axcb = fig.add_axes([0.33, 0.05, 0.33, 0.02])
clb = fig.colorbar(im, cax=axcb, orientation='horizontal')
clb.set_label('Number of Objects')
ax8.set_xlabel(r'Log$_{10}$ (Eddington Ratio $\lambda$)')
ax8.set_ylim(ax2.get_ylim())
ax8.set_xlim(ax7.get_xlim())
plt.tick_params(axis='both', which='major')
ax8.set_yticklabels([])
ax8.set_xticklabels(['', '-1.5', '-1.0', '-0.5', '0.0', '0.5'])
ax7.set_yticklabels([])
ax7.set_xticklabels([])
ax9.hist(tab['LOGBH'][tab['LOGBH'] > 6], color=cset[-1], bins=20)
ax9.set_xlim(ax8.get_xlim())
ax9.set_axis_off()
ax10.hist(tab['RATIO_IR_UV'],
orientation='horizontal',
color=cset[-1],
bins=np.arange(0, 2, 0.1))
ax10.set_ylim(ax1.get_ylim())
ax10.set_axis_off()
ax11.hist(tab['BBT'], orientation='horizontal', color=cset[-1], bins=20)
ax11.set_ylim(ax2.get_ylim())
ax11.set_axis_off()
s1 = spearmanr(tab[tab['LOGEDD_RATIO'] > -2.5]['LOGEDD_RATIO'],
tab[tab['LOGEDD_RATIO'] > -2.5]['RATIO_IR_UV'])[0]
s2 = spearmanr(tab[tab['LOGBH'] > 6]['LOGBH'],
tab[tab['LOGBH'] > 6]['RATIO_IR_UV'])[0]
s3 = spearmanr(tab['LUM_UV'], tab['RATIO_IR_UV'])[0]
s4 = spearmanr(tab[tab['LOGEDD_RATIO'] > -2.5]['LOGEDD_RATIO'],
tab[tab['LOGEDD_RATIO'] > -2.5]['BBT'])[0]
s5 = spearmanr(tab[tab['LOGBH'] > 6]['LOGBH'],
tab[tab['LOGBH'] > 6]['BBT'])[0]
s6 = spearmanr(tab['LUM_UV'], tab['BBT'])[0]
ax1.text(46.5, 0.2, r'$\rho =$ {0:.2f}'.format(s3))
ax4.text(9.7, 0.2, r'$\rho =$ {0:.2f}'.format(s2))
ax7.text(-0.1, 0.2, r'$\rho =$ {0:.2f}'.format(s1))
ax2.text(46.5, 1900, r'$\rho =$ {0:.2f}'.format(s6))
ax5.text(9.7, 1900, r'$\rho =$ {0:.2f}'.format(s5))
ax8.text(-0.1, 1900, r'$\rho =$ {0:.2f}'.format(s4))
fig.savefig('/home/lc585/thesis/figures/chapter06/correlations.pdf')
plt.show()
return None
def main():
#usage: python <thisFile> level <outfile> <file.taxonomy> ..
if len(sys.argv) < 3:
mes = 'usage: python {} level <outfile> <file.taxonomy>..'
print >> sys.stderr, mes.format(os.path.basename(sys.argv[0]))
print >> sys.stderr, "*** filename.split('.')[0] will "\
'be the sample label'
sys.exit(1)
level = int(sys.argv[1])
level = level - 1
outfile = sys.argv[2]
d = {}
dCombined = {}
lisSampOrder = []
for f in sys.argv[3:]:
samp = os.path.basename(f).split('.')[0] # sample name
container, num_lis = readData(f)
tranLis = itertools.izip_longest(*container, fillvalue='Unclassified')
levelLis = list(tranLis)[level]
countD = {}
for tax, num in zip(levelLis, num_lis):
countD[tax] = countD.get(tax, 0) + num
total = sum(countD.values())
d[samp] = dict((taxa, countD[taxa] * 1.0 / total) for taxa in countD)
for key in d[samp]:
dCombined[key] = dCombined.get(key, 0) + d[samp][key]
lisSampOrder.append(samp)
items = sorted(dCombined.items(), key=itemgetter(1), reverse=True)
items = items[:TOP]
if ORDER:
items.reverse()
taxas, nums = zip(*items)
d2 = {}
with open('{}.taxa.plot.summary'.format(outfile), 'wb') as fw2:
print >> fw2, 'Sample\t{}'.format('\t'.join(taxas))
for key in lisSampOrder:
lis = []
for word, count in items[:TOP]:
pct = d[key].get(word, 0)
lis.append(pct * 100)
print >> fw2, '{}\t{}'.format(
key, '\t'.join(('{:.1f}'.format(x) for x in lis)))
d2[key] = lis
sampNum = len(d2)
#color = brewer2mpl.get_map('Set3', 'qualitative', TOP).mpl_colors
color = brewer2mpl.get_map('Paired', 'qualitative', TOP).mpl_colors
almost_black = '#262626'
fig = plt.figure()
ax = fig.add_subplot(111)
margin = 0.1
#width = (1.-2.*margin)/sampNum
width = 0.25
xs = numpy.arange(sampNum)
xs = xs + margin
rects = []
labels = []
liss = []
for samp in lisSampOrder:
labels.append(samp)
liss.append(d2[samp])
#labels, liss = zip(*d2.items())
taxaLiss = zip(*liss)
taxaNum = len(taxaLiss)
assert taxaNum <= TOP
rect = ax.bar(xs, taxaLiss[0], width, color=color[0], linewidth=0)
#rect = ax.bar(xs, taxaLiss[0], width, color=color[0], edgecolor='none')
rects.append(rect)
bottom = numpy.cumsum(taxaLiss, axis=0)
for i in range(1, taxaNum):
rect = ax.bar(xs,
taxaLiss[i],
width,
bottom=bottom[i - 1],
color=color[i],
linewidth=0)
#rect = ax.bar(xs, taxaLiss[i], width, bottom=bottom[i-1], color=color[i], edgecolor='none')
rects.append(rect)
ax.set_ylabel('Percent of total community')
ax.set_xticks(xs + width / 2)
ax.set_xticklabels(labels)
# Remove top and right axes lines ("spines")
spines_to_remove = ['top', 'right']
for spine in spines_to_remove:
ax.spines[spine].set_visible(False)
# Get rid of ticks. The position of the numbers is informative enough of
# the position of the value.
ax.get_xaxis().tick_bottom()
ax.get_yaxis().tick_left()
ax.xaxis.set_ticks_position('none')
#ax.yaxis.set_ticks_position('none')
# For remaining spines, thin out their line and change the black to a slightly off-black dark grey
spines_to_keep = ['bottom', 'left']
for spine in spines_to_keep:
ax.spines[spine].set_linewidth(0.5)
ax.spines[spine].set_color(almost_black)
# shrink current axis by 20%
box = ax.get_position()
ax.set_position([box.x0, box.y0, box.width * 0.6, box.height])
# Change the labels to the off-black
ax.xaxis.label.set_color(almost_black)
ax.yaxis.label.set_color(almost_black)
# Change the axis title to off-black
ax.title.set_color(almost_black)
# Remove the line around the legend box, and instead fill it with a light grey
# Also only use one point for the scatterplot legend because the user will
# get the idea after just one, they don't need three.
#light_grey = np.array([float(248)/float(255)]*3)
legend = ax.legend(rects[::-1],
taxas[::-1],
frameon=True,
loc='center left',
bbox_to_anchor=(1, 0.5))
#legend = ax.legend(lines, keys, loc = 'center left', bbox_to_anchor=(1,0.5))
#legend = ax.legend(frameon=True)
rect = legend.get_frame()
#rect.set_facecolor(light_grey)
rect.set_linewidth(0.0)
# Change the legend label colors to almost black, too
texts = legend.texts
for t in texts:
t.set_color(almost_black)
t.set_fontsize('medium')
t.set_weight('bold')
ax.set_ylim((0, 100.8))
### stuff for log scale
#ax.set_yscale('symlog', linthreshy=0.1)
#yticks = (0.1, 1, 10, 100)
#yticks_str = [str(i) for i in yticks]
#ax.set_yticks(yticks)
#ax.set_yticklabels(yticks_str)
# handling the ticks
#majorLocator = AutoLocator()
#minorLocator = MultipleLocator(10)
#minorLocator = LogLocator(subs=[2,4,6,8])
#minorLocator = AutoMinorLocator()
#minorLocator = LinearLocator()
#minorLocator = MaxNLocator(nbins=9)
#ax.yaxis.set_major_locator(majorLocator)
#ax.yaxis.set_minor_locator(minorLocator)
plt.savefig('{}.png'.format(outfile))
class AreaAggregate(Graph):
"""
This graph will show the combined forested area by summing all countries
together into one graph.
"""
# Size #
height = 7
width = 10
# Basic params #
short_name = 'area_aggregate'
y_grid = True
x_label = "Years"
y_label = "Area in million hectares"
# Colors #
colors = brewer2mpl.get_map('Set1', 'qualitative', 7).mpl_colors
# Name mapping #
label_to_color = {'IPCC': colors[0],
'SOEF': colors[1],
'FAOSTAT': colors[3],
'FRA': colors[4]}
def line_plot(self, axes, x, y, source, **kw):
# Filter by source #
data = area_agg_data.df.query("source == '%s'" % source)
# Plot #
axes.plot(data[x], data[y],
marker = ".",
markersize = 10.0,
color = self.label_to_color[source.upper()],
**kw)
def add_main_legend(self, axes):
items = self.label_to_color.items()
patches = [matplotlib.patches.Patch(color=v, label=k) for k,v in items]
axes.legend(handles = patches,
borderpad = 1,
prop = {'size': 12},
frameon = True,
shadow = True,
loc = 'center left',
bbox_to_anchor = (1.03, 0.5))
def plot(self, **kwargs):
# Plot #
fig = pyplot.figure()
axes = fig.add_subplot(111)
# Plot every data source #
self.line_plot(axes, 'year', 'area', 'ipcc')
self.line_plot(axes, 'year', 'area', 'soef')
self.line_plot(axes, 'year', 'area', 'faostat')
self.line_plot(axes, 'year', 'area', 'fra')
# Leave space for the legend #
fig.subplots_adjust(left=0.1, right=0.8, top=0.95)
# Add legend #
self.add_main_legend(axes)
# Save #
self.save_plot(**kwargs)
# Return for display in notebooks for instance #
return fig
def main():
fdir = "data/processed"
df_flux = pd.read_csv(os.path.join(fdir, "g1_fluxnet_screened_PM.csv"))
df_flux_coupled = pd.read_csv(
os.path.join(fdir, "g1_fluxnet_screened_matching_sites.csv"))
df_leaf = pd.read_csv(os.path.join(fdir, "g1_leaf_gas_exchange.csv"))
df_isotope = pd.read_csv(os.path.join(fdir, "g1_isotope_screened.csv"))
df_sig = pd.read_csv(os.path.join(fdir, "g1_signifletters_withinPFT.csv"))
df_flux['Scale'] = 'FLUXNET - PM'
df_flux_coupled['Scale'] = 'FLUXNET'
df_isotope['Scale'] = 'Leaf isotope'
df = pd.concat([df_flux_coupled, df_leaf, df_flux])
# arbitary cut off of values < 18
aa = len(df)
df = df[df.g1 <= 18]
bb = len(df)
print(aa - bb) / float(aa) * 100.
pfts = [
'ENF', 'EBF', 'DBF', 'TRF', 'SAV', 'SHB', 'C3G', 'C4G', 'C3C', 'C4C'
]
sns.set_style("ticks")
sns.set_style({"xtick.direction": "in", "ytick.direction": "in"})
set2 = brewer2mpl.get_map('Set2', 'qualitative', 8).mpl_colors
fig = plt.figure(figsize=(12, 6))
fig.subplots_adjust(hspace=0.1)
fig.subplots_adjust(wspace=0.05)
plt.rcParams['text.usetex'] = False
plt.rcParams['font.family'] = "sans-serif"
plt.rcParams['font.sans-serif'] = "Helvetica"
plt.rcParams['axes.labelsize'] = 14
plt.rcParams['font.size'] = 12
plt.rcParams['legend.fontsize'] = 12
plt.rcParams['xtick.labelsize'] = 12
plt.rcParams['ytick.labelsize'] = 12
almost_black = '#262626'
# change the tick colors also to the almost black
plt.rcParams['ytick.color'] = almost_black
plt.rcParams['xtick.color'] = almost_black
# change the text colors also to the almost black
plt.rcParams['text.color'] = almost_black
# Change the default axis colors from black to a slightly lighter black,
# and a little thinner (0.5 instead of 1)
plt.rcParams['axes.edgecolor'] = almost_black
plt.rcParams['axes.labelcolor'] = almost_black
ax = fig.add_subplot(111)
# define outlier properties
flierprops = dict(marker='o', markersize=3, markerfacecolor="black")
ax = sns.boxplot(
x="PFT",
y="g1",
hue="Scale",
data=df,
palette=["#66c2a5", "#e78ac3", "#8da0cb"],
order=pfts,
flierprops=flierprops,
hue_order=['Leaf gas exchange', 'FLUXNET - PM', 'FLUXNET'])
plt.legend(loc="upper right")
ax.set_ylabel("$g_1$ (kPa$^{0.5}$)")
ax.set_xlabel("Plant functional type")
#p.set_xlabels(" ")
#p.despine(right=False,top=False,left=False,bottom=False)
ax.set_ylim(-1.5, 14)
ax.set_yticks(np.arange(0, 16, 2))
ax.set_xticklabels(pfts)
for i, p in enumerate(pfts):
if i == 0:
offset = 0
if p != "C4C":
plt.text(-0.13 + offset,
15,
"n=%d\n" % (len(df_leaf[df_leaf.PFT == p].g1)),
horizontalalignment='left',
size='small',
color="#66c2a5",
weight="bold")
if p != "C4G" and p != "C3C" and p != "C4C":
plt.text(-0.13+offset, 14.5,
"n=%d\n" % \
(len(df_flux_coupled[df_flux_coupled.PFT == p].g1)),
horizontalalignment='left', size='small', color="#e78ac3",
weight="bold")
if p != "C4G":
plt.text(-0.13 + offset,
14,
"n=%d\n" % (len(df_flux[df_flux.PFT == p].g1)),
horizontalalignment='left',
size='small',
color="#8da0cb",
weight="bold")
offset += 1.0
odir = "/Users/%s/Dropbox/g1_leaf_ecosystem_paper/figures/figs/" %\
(os.getlogin())
#odir = "/Users/%s/Desktop/" % (os.getlogin())
plt.savefig(os.path.join(
odir, "g1_gas_exchange_isotope_fluxnet_boxplot_inverted_PM.pdf"),
bbox_inches='tight',
pad_inches=0.1)
for line in data:
toks = line.strip().split(',')
power_budget.append(toks[0])
hocmp_value.append(toks[1])
hecmp_value.append(toks[2])
improvement.append(((float(toks[2]) / float(toks[1])) - 1) * 100)
fig, ax1 = plt.subplots()
num_plots = 3
###Create markers
#markers = itertools.cycle(('+', '.', 'o', '*', '^'))
markers = itertools.cycle(('o', '*', '^'))
##Color Map
set2 = brewer2mpl.get_map('Set1', 'qualitative', 9).mpl_colors
t = 1
ax1.plot(power_budget,
improvement,
color=set2[t],
label=sys.argv[2],
marker=markers.next(),
markersize=4,
linewidth=2)
ax1.set_xlabel('Power Budget (W)')
# Make the y-axis label and tick labels match the line color.
ax1.set_ylabel('Improvement(%)', color=set2[t])
for tl in ax1.get_yticklabels():
tl.set_color(set2[t])
import os
#Science packages
import matplotlib.pyplot as plt
from matplotlib.dates import DateFormatter
import matplotlib.ticker as plticker
import numpy as np
from scipy import stats
#Other Packages
import brewer2mpl as b2m #colorbrewer maps
"""----------------------------- plot setup ------------------------------------------"""
#set up color map with brewer2mpl, not mandatory as matplotlib as the colorbrewer maps but
# provides an easy way to limit the number of colors in a map
bmap = b2m.get_map('Blues', 'Sequential', 5, reverse=False)
bmap = bmap.mpl_colors
"""----------------------------- plots ----------------------------------------------"""
def plot_wavetransf(wa, T, S, sig95, time_base, plot_percentile=False):
"""plotting WaveTransform Power with confidence interval contour"""
fig = plt.figure(1)
ax = plt.subplot(1, 1, 1)
if plot_percentile:
#use following to contour at "percentiles variances" when using non-normalized data to match web output
csf = plt.contourf(T,
S,
wa.wavelet_power,
levels=[
def plot_aa(analysis_dir, params, runs, refnames, plot_type, sel='backbone', \
pltparams=(None, None, (5.75, 12)), plotname=None, showplot=True):
#-----------------------
# Timeseries angles for 1ake and 4ake trajectories
#-----------------------
methods, protein, directions = params
drmsd, angles, dmcd, idf = False, True, False, False
st = en = sk = None
N = len(runs)
ofname_ref_nmp, ofname_ref_lid = refnames
timeseries, twod = plot_type
xlims, ylims, (fsize1, fsize2), lsize = pltparams
#-- plot settings ------
xmin, xmax = xlims
ymin, ymax = ylims
xdatarange, ydatarange = (xmax-xmin), (ymax-ymin)
#-----------------------
# 2D angles for 1ake and 4ake trajectories
#-----------------------
if twod:
print "Plotting angles for 1ake and 4ake trajectories..."
# Plot colors:
import brewer2mpl
import seaborn as sns
sns.set_style('whitegrid')
# Save a nice dark grey as a variable
almost_black = '#262626'
# Get "Set2" colors from ColorBrewer (all colorbrewer scales: http://bl.ocks.org/mbostock/5577023)
dark2 = brewer2mpl.get_map('Dark2', 'qualitative', 8).mpl_colors
accent = brewer2mpl.get_map('Accent', 'qualitative', 8).mpl_colors
set1 = brewer2mpl.get_map('Set1', 'qualitative', 9).mpl_colors
s2 = brewer2mpl.get_map('Set2', 'qualitative', 8).mpl_colors
s3 = brewer2mpl.get_map('Set3', 'qualitative', 12).mpl_colors
# colors = [s3[9],s3[5],s2[4],s3[3],s3[7],s2[5],s2[7],s2[6],s2[0],s3[10],s2[1]]
# colors = [s3[0],s3[11]] + s3[2:8] + s3[9:11] + [s2[6]]
# colors = [s3[1],s2[6]]+s3[2:8] + [s2[7]] + s3[9:11] +[s3[8]]
# colors = [s3[11],s2[6],s3[2],s3[3],s2[0],s3[5],s3[7],s3[9],s3[4],s2[4],'gray']
# colors = [s2[0],s3[11],s3[2],s3[3],s3[4],s3[5],s2[4],s3[7],s2[6],s3[9],'gray']
# colors = [s2[0],s3[11],s2[3],s3[3],s2[2],s3[7],s2[4],s3[5],s3[9],s2[6],'gray']
# xkcd_colors = ["pinkish", "brick red",
# "goldenrod", "orange",
# "buff", "banana yellow",
# "pale lime green", "tree green",
# "robin egg blue", "windows blue",
# "lilac", "purplish",
# "tan", "raw umber",
# "grey"]
xkcd_colors = ["pinkish", "brick red", "goldenrod", "orange",
"banana yellow", "pale lime green", "tree green",
"robin egg blue", "windows blue", "lilac", "purplish",
"raw umber", "grey"]
xkcd_colors = ["cherry", "tree green", "orange", "windows blue",
"banana yellow", "bubblegum pink", "goldenrod",
"robin egg blue", "pale lime green", "lilac", "purply",
"raw umber", "grey"]
colors = sns.xkcd_palette(xkcd_colors)
end_colors = sns.xkcd_palette(['vibrant green', 'bright red'])
# Turn off lines and multiple markers in legend
mpl.rcParams['legend.handlelength'] = 0
mpl.rcParams['legend.numpoints'] = 1
mpl.rcParams['legend.scatterpoints'] = 1
plt.rcParams['xtick.major.pad']='3'
plt.rcParams['ytick.major.pad']='3'
#-- create figure ------
plt.close("all")
aspect_ratio = float(xdatarange)/ydatarange
fig = plt.figure(1, figsize=(fsize1, fsize2))
ax = fig.add_subplot(111)
if sel == 'name CA':
postf = '_ca'
elif sel == 'backbone':
postf = '_bb'
ref_theta_nmp = np.loadtxt(ofname_ref_nmp+postf+'.dat')
ref_theta_lid = np.loadtxt(ofname_ref_lid+postf+'.dat')
#-- plot reference structures ----------
s1 = ax.scatter(ref_theta_nmp[0], ref_theta_lid[0], marker="o", s=100, \
label=" 1AKE", c=end_colors[0], zorder=15, linewidths=1)
# s1 = ax.scatter(ref_theta_nmp[0], ref_theta_lid[0], marker="o", s=150, \
# c='gray', zorder=14, linewidths=0)
s2 = ax.scatter(ref_theta_nmp[1], ref_theta_lid[1], marker="D", s=90, \
label=" 4AKE", c=end_colors[1], zorder=15, linewidths=1)
# s2 = ax.scatter(ref_theta_nmp[1], ref_theta_lid[1], marker="D", s=140, \
# c='gray', zorder=14, linewidths=0)
for i, m in zip(xrange(len(methods)), methods):
method, methodname, postfix = m[0], m[1], m[2]
d = 'co'
theta_nmp_accum = []
theta_lid_accum = []
for run in xrange(N):
mn = ' %s' % methodname
mtype = 'o' if d=='co' else 'D'
if run > 0:
mn = None
if (method == 'intp'): break # only one interp line traj
if (method == 'intp' and d == 'oc'): break
if postfix != None:
basename = '%s/%s_%s-%s-%s-%s' \
% (analysis_dir, 'aa', method, postfix, d, protein)
else:
basename = '%s/%s_%s-%s-%s' \
% (analysis_dir, 'aa', method, d, protein)
### Hack to use all-atom intp trajectories ('_full' vs '_ca')
if method == 'intp':
ofname1 = '%s-%s_%s_full.dat' % (basename, 'nmp', runs[run])
ofname2 = '%s-%s_%s_full.dat' % (basename, 'lid', runs[run])
else:
ofname1 = '%s-%s_%s.dat' % (basename, 'nmp', runs[run])
ofname2 = '%s-%s_%s.dat' % (basename, 'lid', runs[run])
sk = 2 if method == 'map' else 1 # reduce # map frames
theta_trj_nmp = np.loadtxt(ofname1)[st:en:sk]
theta_trj_lid = np.loadtxt(ofname2)[st:en:sk]
theta_nmp_accum = np.append(theta_nmp_accum, theta_trj_nmp)
theta_lid_accum = np.append(theta_lid_accum, theta_trj_lid)
color = colors[i]
zorder = 12-i
(mtype,size) = ('D',15) if i < 6 else ('o',21)
ax.plot(theta_trj_nmp, theta_trj_lid, lw=1, \
c=color, alpha=1, zorder=zorder-1)
# zorder = -i #if 'menm' not in method else 9
ax.scatter(theta_trj_nmp, theta_trj_lid, lw=0.2, s=size, \
c=color, marker=mtype, label=mn, alpha=1, zorder=zorder)
#----------------------------------------
# Legends
#----------------------------------------
# Thin line around the legend box, and instead fill it with a light grey
# Also only use one point for the scatterplot legend because the user will
# get the idea after just one, they don't need three.
# axl = fig.add_subplot(122)
# light_grey = np.array([float(248)/float(255)]*3)
# legend = axl.legend(loc='lower left', bbox_to_anchor=(0.85, 0.025), \
# frameon=True, fancybox=True, scatterpoints=1, borderpad=1, \
# labelspacing=0.5, markerscale=1.0)
# rect = legend.get_frame()
# rect.set_facecolor(light_grey)
# rect.set_linewidth(1.0)
# Change the legend label colors to almost black, too
# for t in legend.texts: t.set_color(almost_black)
# plt.setp(legend.get_texts(), fontsize=(lsize-2))
#----------------------------------------
legend = None
ax.set_xlim([xmin,xmax])
ax.set_ylim([ymin,ymax])
# Remove top and right axes lines ("spines")
# spines_to_remove = ['top', 'right']
# for spine in spines_to_remove:
# ax.spines[spine].set_visible(False)
# Locate major ticks at 5 deg intervals on both axes
intvl = 10
x_major_ticklocs = np.arange(40, 75+intvl, intvl)
y_major_ticklocs = np.arange(100, 155+intvl, intvl)
locx = plticker.FixedLocator(x_major_ticklocs)
locy = plticker.FixedLocator(y_major_ticklocs)
ax.xaxis.set_major_locator(locx)
ax.yaxis.set_major_locator(locy)
ax.tick_params(axis='both', which='major', labelsize=lsize, pad=-7.5)
# major_formatter = plticker.FormatStrFormatter(r'%3i$^\circ$')
# ax.xaxis.set_major_formatter(major_formatter)
# ax.yaxis.set_major_formatter(major_formatter)
xticklabels = np.array(ax.get_xticks())
yticklabels = np.array(ax.get_yticks())
# x_major_ticklocs = np.arange(xmin, xmax+intvl, intvl)
xticklabels = ['' if (t%2).round() != 0 else r'%3i$^\circ$' % t for t in xticklabels] # change fraction to %
yticklabels = ['' if (t%2).round() != 0 else r'%3i$^\circ$' % t for t in yticklabels] # change fraction to %
ax.set_xticklabels(xticklabels, minor=False)
ax.set_yticklabels(yticklabels, minor=False)
# ax.grid(False)
# ax.tick_params(axis='both', which='major', labelsize=lsize)
# ax.grid(True, which='major', ls='-', c=almost_black, alpha=1)
# rec = mpatches.Rectangle((xmin,ymin), xdatarange, ydatarange, \
# fill=False,lw=1, color='grey', alpha=0.4)
# rec = ax.add_patch(rec)
# rec.set_clip_on(False)
#-- labels ------
ax.set_xlabel(r'NMP$-$CORE angle', fontsize=lsize)
ax.set_ylabel(r'LID$-$CORE angle', fontsize=lsize)
# ax.xaxis.set_ticks_position('none')
# ax.yaxis.set_ticks_position('none')
sns.despine(offset=10, bottom=True, left=True)
plt.tight_layout()
def multisavefig(name, lgd=None, transparent=True):
for ext in ['tif', 'pdf', 'png']:
if lgd is None:
plt.savefig(name+'.'+ext, format=ext, \
transparent=transparent, bbox_inches='tight')
else:
plt.savefig(name+'.'+ext, format=ext, \
transparent=transparent, \
bbox_extra_artists=(lgd,), bbox_inches='tight')
if (plotname is not None):
if type(plotname) is str:
pltfp1, pltfp2 = plotname, None
else: pltfp1, pltfp2 = plotname
# multisavefig(pltfp1, lgd=legend, transparent=True)
multisavefig(pltfp1, transparent=True)
# multisavefig(pltfp1+'-notrans', lgd=legend, transparent=False)
if pltfp2 is not None:
import glob, os, shutil, re
source_dir_split = [x for x in pltfp1.split('/')]
source_dir = '/'.join(source_dir_split[:-1])
name = '%s*' % pltfp1
files = glob.iglob(os.path.join(source_dir, name))
for f in files:
if os.path.isfile(f):
ext = f.split('.')[-1]
newfile = '%s.%s' % (pltfp2, ext)
newfile_t = '%s-notrans.%s' % (pltfp2, ext)
shutil.copyfile(f, newfile)
shutil.copyfile(f, newfile_t)
if showplot:
plt.show()
plt.close()
reference_genomes = OrderedDict([
('ont_lambda', '/data/minion/reference_genomes/lambda/lambda_ref.fasta'),
('cs', '/data/minion/reference_genomes/cs/dna_cs.fasta'),
('3D7_V3', '/data/minion/reference_genomes/3D7_V3/3D7_V3.fasta'),
('AgamP3', '/data/minion/reference_genomes/AgamP3/Anopheles-gambiae-PEST_CHROMOSOMES_AgamP3.fasta')
])
# <codecell>
read_types = ['template', 'complement', '2D']
# <codecell>
import brewer2mpl
set1 = brewer2mpl.get_map('Set1', 'qualitative', 8).mpl_colors
set3 = brewer2mpl.get_map('Set3', 'qualitative', 12).mpl_colors
# <headingcell level=1>
# Functions
# <codecell>
ids = OrderedDict([
('2D basecalling successful' , 'Basecalling 2D hairpin data'),
('Missing Complement data' , 'Splitter indicates no complement data. Workflow cannot continue'),
('Missing Template data' , 'Splitter indicates no template data. Workflow cannot continue'),
('Too few events' , 'Read has too few events. Workflow cannot continue'),
('Poor ratio of template to complement data' , 'Ratio of template to complement data is no good. Workflow cannot continue'),
('Inconsistent template vs complement events' , 'Number of template events out of range aligning template'),
topo_fname = os.path.join(os.getenv('SARIKA_INPUT'), 'TOPO-124x124.nc')
wrf_fname = os.path.join(os.getenv('SARIKA_INPUT'),
'wrfheight-124x124-22levs.nc')
bd = sp.parse_STEM_var(bd_fname,
varname='CO2_TRACER1',
t0=datetime(2008, 7, 1),
t1=datetime(2008, 9, 1))
w_bnd_tavg = np.mean(bd['data'][:, :, -125:-1], axis=0) * 1000
stemlon, stemlat, topo = sp.parse_STEM_coordinates(topo_fname)
agl, asl = get_STEMZ_height(topo_fname, wrf_fname)
cos_norm = midpt_norm.MidpointNormalize(midpoint=450)
cos_cmap = brewer2mpl.get_map('PuOr', 'diverging', 8).mpl_colormap
fig, ax = plt.subplots()
lat = stemlat[:, -1]
ax.set_xlim([lat.min(), lat.max()])
cm = ax.pcolormesh(lat,
agl[:, :, -1],
w_bnd_tavg,
cmap=plt.get_cmap('PuOr'),
norm=cos_norm)
ax.set_xlabel('latitude ($^\circ$N)')
ax.set_ylabel('height AGL (m)')
ax.set_title('PCTM Western boundary July/August mean')
ticks = np.round(
np.concatenate([
np.linspace(w_bnd_tavg.min(), 450, 4),
def generate_plots(df, m, bin_size, min_bins):
# Map-related info
longitudes = np.arange(minLon, maxLon, 0.25)
latitudes = np.arange(minLat, maxLat, 0.1)
#Relevant quatities
categs = pd.Series(sorted(df['PropertyType'].unique()))
n_categs = len(categs)
n_samples = 1. * len(fire)
# Posterior mean estimate of category probabilities using Dirichlet prior.
categ_priors = (df.groupby(['PropertyType']).size() + 1) / (n_samples +
n_categs)
# Create the x and y bin edges over which we'll smooth positions.
# Better approach might be to find optimal bin size through
# cross validation based approach (need more time though).
bin_edges_x = np.arange(
np.min(df['X']) - bin_size / 2,
np.max(df['X']) + bin_size / 2, bin_size)
bin_edges_y = np.arange(
np.min(df['Y']) - bin_size / 2,
np.max(df['Y']) + bin_size / 2, bin_size)
overall_hist, yedges, xedges, Img = plt.hist2d(df.Y.values,
df.X.values,
bins=(bin_edges_y,
bin_edges_x))
# We'll assume that fires really only occurs at a location
# at which we have seen at least one fire incidence over
#the period comprising 2012-2015.
# This will help with the plotting
mask = overall_hist == 0
n_bins = np.sum(overall_hist > 0)
# Obtain the class condition probabilities p(x|y).
# Essentially we are computing the quantity: given the type of affected property,
# what is the probability that the fire occurred in a given xy bin.
# Because a single fire can happen in only one location, we are
# treating the class conditional densities as multinomial.
groups = df.groupby(['PropertyType'])
px_y = np.zeros([len(bin_edges_y) - 1, len(bin_edges_x) - 1, n_categs])
px_y_ma = np.ma.masked_where(
np.tile(np.atleast_3d(mask), [1, 1, n_categs]), px_y)
for i, (name, group) in enumerate(groups):
group_hist, yedges, xedges, Img = plt.hist2d(group.Y.values,
group.X.values,
bins=(bin_edges_y,
bin_edges_x))
group_hist_ma = np.ma.masked_where(mask, group_hist)
# Posterior mean estimates of class conditonal probabilities
# using Dirichlet prior.
px_y_ma[:, :,
i] = (group_hist_ma + 1.0) / (np.sum(group_hist_ma) + n_bins)
# Present the category prior into the right shape
p_y = np.atleast_3d(categ_priors.as_matrix()).reshape(1, 1, n_categs)
p_y = np.tile(p_y, [len(bin_edges_y) - 1, len(bin_edges_x) - 1, 1])
p_y_ma = np.ma.masked_where(np.tile(np.atleast_3d(mask), [1, 1, n_categs]),
p_y)
# Obtain the posterior probabilites of each property type,
# given that the fire occurred in a known xy location
py_x_ma = (p_y_ma * px_y_ma) / np.atleast_3d(
np.sum(p_y_ma * px_y_ma, axis=2))
# Compute entropy of posterior distribution
hy = -np.sum(categ_priors * np.log2(categ_priors))
hy_x = -np.sum(py_x_ma * np.log2(py_x_ma), axis=2)
entropy_diff = hy - hy_x
X, Y = np.meshgrid(bin_edges_x, bin_edges_y)
# Get the property type that maximized the posterior probability.
winner = np.argmax(py_x_ma, axis=2)
winner_ma = np.ma.masked_where(mask, winner)
# How many time does each fire(property type) prove to be the most likely?
counts = itemfreq(winner_ma[winner_ma.mask == False])
# Sort the counts and take only those types which have
# at least min_pop_bins bins in which they were the winner
sorted_counts = counts[np.argsort(counts[:, 1])[::-1], :]
top_counts = sorted_counts[sorted_counts[:, 1] >= min_bins, :]
#top_counts_less = top_counts[1:, :]
top_counts_less = top_counts
####### Plotting preferences
plt.figure(figsize=(16, 18))
m.arcgisimage(service='Canvas/World_Light_Gray_Base',
xpixels=1200,
ypixels=1000)
#plt.figure(figsize=(16,18))
#m.drawmapboundary(fill_color='paleturquoise')
#m.fillcontinents(color=(0.25, 0.25, 0.25), zorder=0)
#m.drawparallels(latitudes, color='white', labels=[1, 0, 0, 0],
# dashes=[5, 5], linewidth=0.75)
#m.drawmeridians(longitudes, color='white', labels=[0, 0, 0, 1],
# dashes=[5, 5], linewidth=0.75)
n_colors = top_counts_less.shape[0]
colors = b2mpl.get_map('Set3', 'Qualitative', n_colors).mpl_colors
recoded = np.zeros(winner.shape)
for i, categ_num in enumerate(top_counts[:, 0]):
recoded[winner == categ_num] = i
winner_ma = np.ma.masked_where(recoded == 0, recoded)
m.pcolormesh(X,
Y,
winner_ma,
cmap=mpl_colors.ListedColormap(colors),
latlon=True)
winner_ma = np.ma.masked_where(np.logical_or(mask, winner != 16), winner)
m.pcolormesh(X, Y, winner_ma, cmap='Greys', alpha=0.5, edgecolor='None')
legend_labels = [categs[i] for i in top_counts[:, 0]]
# separate way of doing legend for the most common crime, which is not
# part of the colormap
legend_markers = []
legend_markers.append(Rectangle((0, 0), 1, 1, fc='thistle', ec='thistle'))
# now the other types
legend_edgecolors = colors
legend_facecolors = colors
for i in range(len(top_counts_less)):
legend_markers.append(
Rectangle((0, 0),
1,
1,
fc=legend_facecolors[i],
ec=legend_edgecolors[i]))
legend = plt.legend(legend_markers,
legend_labels,
labelspacing=.075,
handlelength=.5,
handletextpad=.1,
fancybox=True,
frameon=1,
loc='lower left')
frame = legend.get_frame()
frame.set_facecolor('black')
texts = legend.get_texts()
texts[0].set_fontsize(10)
texts[0].set_color('white')
for t, c in zip(texts[1:], legend_edgecolors):
t.set_fontsize(10.5)
t.set_color(c)
plt.title(
"Most likely fire (specified by affected property type) \n"
"given knowledge of location only,\n"
"taking into account the prior fire probability\n\n",
fontsize=14)
plt.savefig('Type_output.png')
#### Entropy difference plot
plt.figure(figsize=(16, 18))
entropy_diff_ma = np.ma.masked_where(mask, entropy_diff)
max_diff = np.max(entropy_diff_ma)
min_diff = np.min(entropy_diff_ma)
max_abs = np.max([max_diff, np.abs(min_diff)])
m.pcolormesh(X,
Y,
entropy_diff_ma,
cmap='BuPu',
alpha=0.75,
edgecolor='None',
vmin=-max_abs,
vmax=max_abs)
m.arcgisimage(service='Canvas/World_Light_Gray_Base', xpixels=1500)
#m.drawmapboundary(fill_color='paleturquoise')
#m.fillcontinents(color=(0.25, 0.25, 0.25), zorder=0)
#m.drawparallels(latitudes, color='white', labels=[1, 0, 0, 0],
# dashes=[5, 5], linewidth=.75)
#m.drawmeridians(longitudes, color='white', labels=[0, 0, 0, 1],
# dashes=[5, 5], linewidth=.75)
m.scatter(fire.X, fire.Y)
m.pcolormesh(X,
Y,
entropy_diff_ma,
cmap='BuPu',
alpha=0.75,
edgecolor='None',
vmin=-max_abs,
vmax=max_abs)
cbar = plt.colorbar(shrink=0.25)
cbar.solids.set_edgecolor("face")
plt.title(
'Entropy of prior distribution minus entropy of posterior distribution:\n'
'positive values indicate less uncertainty about fire type '
'after observing location\n\n',
fontsize=14)
plt.savefig('Entropy.png')
class SimulationRunner(object):
Simulation = namedtuple('Simulation', [
'predictor_config', 'feature_cardinality', 'num_examples',
'biased_feature_proportion', 'directory', 'extension',
'visualization_interval', 'biased_feature_effect_length'
])
COLORS = brewer2mpl.get_map('Set2', 'qualitative', 8).mpl_colors
def __init__(self, simulation):
self._simulation = simulation
self._predictor = AdPredictor(simulation.predictor_config)
self._sampler = Sampler(simulation)
def _current_weights_by_feature(self):
by_feature = lambda kv: kv[0].feature
by_feature_value = lambda kv: (kv[0].feature, kv[0].value)
weights = sorted(self._predictor.weights, key=by_feature_value)
for feature, group in itertools.groupby(weights, key=by_feature):
yield feature, [(f, w.mean, w.variance) for (f, w) in group]
def _plot_weights(self):
for color, (feature, weights) in itertools.izip(
itertools.cycle(self.COLORS),
self._current_weights_by_feature()):
_, means, variances = zip(*weights)
logging.debug("Feature %s, Weights: %s", feature, weights)
label = "F{}".format(feature) if feature != 0 else "Bias"
plt.scatter(means,
variances,
label=label,
color=color,
alpha=0.8,
s=40)
def _annotate_biased_weights(self):
for _, weights in self._current_weights_by_feature():
for (feature, mean, variance) in weights:
bias_weight = self._sampler.get_bias_for_feature(feature)
if bias_weight is not None:
plt.annotate('+' if bias_weight else '-', (mean, variance),
size=40)
def _visualize(self, num_examples):
plt.clf()
self._plot_weights()
self._annotate_biased_weights()
plt.title(u"(μ, σ²) after {} examples".format(num_examples))
plt.xlabel(u"μ")
plt.ylabel(u"σ²")
plt.legend(loc='best')
plt.xlim(-4, 4)
plt.ylim(-0.1, 1.1)
filename = "{:03d}.{}".format(num_examples, self._simulation.extension)
logger.info("Saving graph to %s", filename)
plt.savefig(os.path.join(self._simulation.directory, filename),
dpi=300)
def run(self):
samples = itertools.islice(self._sampler,
self._simulation.num_examples)
for iteration, (features, label) in enumerate(samples):
self._predictor.train(features, label)
if iteration % self._simulation.visualization_interval == 0:
self._visualize(iteration)
import numpy as np
import brewer2mpl
from matplotlib import cm
# ColorBrewer by Drs. Cynthia Brewer and Mark Harrower of Pennsylvania
# State University. For more information on ColorBrewer, see:
# - Flash-based interactive map:
# http://colorbrewer2.org/
# - A quick visual reference to every ColorBrewer scale:
# http://bl.ocks.org/mbostock/5577023
# ColorBrewer scale 'Qualitative.Set1'.
# This one has nice "traditional" colors like reds and blues
brewer_set1 = brewer2mpl.get_map('Set1', 'qualitative', 9).mpl_colors
# Remove the sixth color (yellow) which is too bright
brewer_set1.pop(5)
# Swap the red and blue to let blue come first
brewer_set1[0], brewer_set1[1] = brewer_set1[1], brewer_set1[0]
# ColorBrewer scale 'Qualitative.Set2'.
# This one are less saturated than those colors in Set1.
brewer_set2 = brewer2mpl.get_map('Set2', 'qualitative', 8).mpl_colors
# ColorBrewer scale 'Qualitative.Set3'.
# This one are even less saturated than those colors in Set1.
brewer_set3 = brewer2mpl.get_map('Set3', 'qualitative', 12).mpl_colors
# Remove the second color (yellow) which is too bright
brewer_set3.pop(1)
plt.show()
if __name__ == '__main__':
#manual()
#radii = np.array([5,10,20,30,40,60,80,100,150,200,250])
radii = np.array([5, 20, 40, 60, 80, 100, 150, 200, 250])
#radii = np.array([5,60,200])
#on_pix_um_radius = 50
center_amp_vec = np.zeros_like(radii)
#plot_settings
colors = cb.get_map('YlOrRd', 'Sequential', 9).mpl_colors
plt.figure(figsize=(10, 10))
plt.subplot(2, 1, 1)
for idx, r in enumerate(radii):
dmd = DMDchip(r, chip_side_length_pixels=100)
sim = FourierSim(xrng=[-500.0, 500.0],
yrng=[-500.0, 500.0],
N=[500, 500],
dmd=dmd)
sim.dmd.compute_on_pixels()
sim.compute_initial_amplitude_distribution()
sim.fourier_filter()
rads, avg, center_amp = sim.compute_radial_average()
@author: isaac
"""
import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
import plots
reload(plots)
from plots import bmap, rcj, tl, tickout, four_plot
# %% Get some colors
import brewer2mpl
new_bmap = brewer2mpl.get_map('Set1', 'Qualitative', 9).mpl_colors
new_bmap.pop(5)
more_colors = brewer2mpl.get_map('Set2', 'Qualitative', 8).mpl_colors
new_bmap += more_colors
# %%
d = pd.read_csv('./Data/epsilon.csv')
mass = d[u'Mass (g)'].values
eps = d[u'epsilon'].values
eps_snake = d[u'eps from c from sqrt(Ws)'].values
Ws = d[u'Wing loading (N/m^2)']
all_labels = d[u'Label']
