def perf_std_discrete(ticks,
avgs,
stds,
legend=None,
std_width=0.3,
plot_width=1000,
plot_height=300,
color=BLUE,
alpha=1.0,
**kwargs):
plotting.rect(ticks,
avgs, [std_width for _ in stds],
2 * np.array(stds),
line_color=None,
fill_color=color,
fill_alpha=alpha * 0.5,
plot_width=plot_width,
plot_height=plot_height,
**kwargs)
plotting.hold(True)
plotting.line(ticks,
avgs,
line_color=color,
line_alpha=alpha,
legend=legend)
plotting.circle(ticks,
avgs,
line_color=None,
fill_color=color,
fill_alpha=alpha)
plotting.grid().grid_line_color = 'white'
plotting_axis()
plotting.hold(False)
def mesh(meshgrid, s_vectors=(), s_goals=(),
mesh_timescale=(1000000,), mesh_colors=(C_COLOR_H,), title='no title',
e_radius=1.0, e_color=E_COLOR, e_alpha=0.75,
g_radius=1.0, g_color=G_COLOR, g_alpha=0.75, swap_xy=True, tile_ratio=0.97,
x_range=None, y_range=None):
x_range, y_range = ranges(meshgrid.s_channels, x_range=x_range, y_range=y_range)
xm = zip(*[b.bounds[0] for b in meshgrid.nonempty_bins])
ym = zip(*[b.bounds[1] for b in meshgrid.nonempty_bins])
if swap_xy:
x_range, y_range = y_range, x_range
xm, ym = ym, xm
color = []
for b in meshgrid.nonempty_bins:
t = b.elements[0][0]
color.append(mesh_colors[bisect.bisect_left(mesh_timescale, t)])
plotting.rect((np.array(xm[1])+np.array(xm[0]))/2 , (np.array(ym[1])+np.array(ym[0]))/2,
(np.array(xm[1])-np.array(xm[0]))*tile_ratio, (np.array(ym[1])-np.array(ym[0]))*tile_ratio,
x_range=x_range, y_range=y_range,
fill_color=color, fill_alpha=0.5,
line_color='#444444', line_alpha=0.0, title=title)
plotting.hold(True)
plotting.grid().grid_line_color='white'
spread(meshgrid.s_channels, s_vectors=s_vectors, s_goals=s_goals, swap_xy=swap_xy,
e_radius=e_radius, e_color=e_color, e_alpha=e_alpha,
g_radius=g_radius, g_color=g_color, g_alpha=g_alpha)
plotting.hold(False)
def mesh(meshgrid,
s_vectors=(),
s_goals=(),
mesh_timescale=(1000000, ),
mesh_colors=(C_COLOR_H, ),
title='no title',
e_radius=1.0,
e_color=E_COLOR,
e_alpha=0.75,
g_radius=1.0,
g_color=G_COLOR,
g_alpha=0.75,
swap_xy=True,
tile_ratio=0.97,
x_range=None,
y_range=None):
x_range, y_range = ranges(meshgrid.s_channels,
x_range=x_range,
y_range=y_range)
xm = zip(*[b.bounds[0] for b in meshgrid.nonempty_bins])
ym = zip(*[b.bounds[1] for b in meshgrid.nonempty_bins])
if swap_xy:
x_range, y_range = y_range, x_range
xm, ym = ym, xm
color = []
for b in meshgrid.nonempty_bins:
t = b.elements[0][0]
color.append(mesh_colors[bisect.bisect_left(mesh_timescale, t)])
plotting.rect((np.array(xm[1]) + np.array(xm[0])) / 2,
(np.array(ym[1]) + np.array(ym[0])) / 2,
(np.array(xm[1]) - np.array(xm[0])) * tile_ratio,
(np.array(ym[1]) - np.array(ym[0])) * tile_ratio,
x_range=x_range,
y_range=y_range,
fill_color=color,
fill_alpha=0.5,
line_color='#444444',
line_alpha=0.0,
title=title)
plotting.hold(True)
plotting.grid().grid_line_color = 'white'
spread(meshgrid.s_channels,
s_vectors=s_vectors,
s_goals=s_goals,
swap_xy=swap_xy,
e_radius=e_radius,
e_color=e_color,
e_alpha=e_alpha,
g_radius=g_radius,
g_color=g_color,
g_alpha=g_alpha)
plotting.hold(False)
def line(x_range, avg, std, color='#E84A5F', dashes=(4, 2), alpha=1.0):
plotting.line(x_range, [avg, avg],
line_color=color,
line_dash=list(dashes),
line_alpha=alpha)
plotting.hold(True)
plotting.rect([(x_range[0] + x_range[1]) / 2.0], [avg],
[x_range[1] - x_range[0]], [2 * std],
fill_color=color,
line_color=None,
fill_alpha=0.1 * alpha)
plotting.hold(False)
def build_punchcard(datamap_list, concept_list,
radii_list, fields_in_concept_list,
datamaps, concepts,
plot_width=1200, plot_height=800):
source = ColumnDataSource(
data=dict(
datamap=datamap_list, # x
concept=concept_list, # y
radii=radii_list,
fields_in_concept=fields_in_concept_list,
)
)
output_file('')
hold()
figure()
plot_properties = {
'title': None,
'tools': "hover,resize,previewsave",
'y_range': [get_datamap_label(datamap) for datamap in datamaps],
'x_range': concepts,
'plot_width': plot_width,
'plot_height': plot_height,
}
rect('concept', 'datamap', # x, y
1, 1, # height, width
source=source,
color='white', # put in background
**plot_properties)
circle('concept', 'datamap', # x, y
size='radii',
source=source,
color='black',
**plot_properties)
grid().grid_line_color = None
x = xaxis()
x.major_label_orientation = pi / 4
hover = [t for t in curplot().tools if isinstance(t, HoverTool)][0]
hover.tooltips = OrderedDict([
("Datamap", "@datamap"),
("Concept", "@concept"),
("Fields", "@fields_in_concept"),
])
return curplot().create_html_snippet(
static_path=settings.STATIC_URL,
embed_save_loc=settings.BOKEH_EMBED_JS_DIR,
embed_base_url=reverse('bokeh'),
)
def perf_std_discrete(ticks, avgs, stds, legend=None,
std_width=0.3, plot_width=1000, plot_height=300,
color=BLUE, alpha=1.0, **kwargs):
plotting.rect(ticks, avgs, [std_width for _ in stds], 2*np.array(stds),
line_color=None, fill_color=color, fill_alpha=alpha*0.5,
plot_width=plot_width, plot_height=plot_height, **kwargs)
plotting.hold(True)
plotting.line(ticks, avgs, line_color=color, line_alpha=alpha, legend=legend)
plotting.circle(ticks, avgs, line_color=None, fill_color=color, fill_alpha=alpha)
plotting.grid().grid_line_color = 'white'
plotting_axis()
plotting.hold(False)
def line(x_range, avg, std, color="#E84A5F", dashes=(4, 2), alpha=1.0):
plotting.line(x_range, [avg, avg], line_color=color, line_dash=list(dashes), line_alpha=alpha)
plotting.hold(True)
plotting.rect(
[(x_range[0] + x_range[1]) / 2.0],
[avg],
[x_range[1] - x_range[0]],
[2 * std],
fill_color=color,
line_color=None,
fill_alpha=0.1 * alpha,
)
plotting.hold(False)
def make_box_violin_plot_2(data, maxwidth=0.9):
"""
data: dict[Str -> List[Number]]
maxwidth: float
Maximum width of tornado plot within each factor/facet
Returns the plot object
"""
df = pd.DataFrame(columns=["group", "centers", "width", "height", "texts"])
bar_height = 50
bins = [0, 1e1, 1e2, 1e3, 1e4, 1e5, 1e6]
# Compute histograms, while keeping track of max Y values and max counts
for i, (group, vals) in enumerate(data.iteritems()):
hist, edges = np.histogram(vals, bins)
df = df.append(pd.DataFrame(dict(
group = group,
centers = np.arange(len(hist))*bar_height,
width = np.log10(hist),
height = np.ones(hist.shape)*bar_height,
texts = map(str,hist),
)))
df.replace(-np.inf, 0)
# Normalize the widths
df["width"] *= (maxwidth / df["width"].max())
hold()
width, height = 800, 800
figure(plot_width=width, plot_height=height, title="Degree Distribution",
tools="previewsave",
x_axis_type=None, y_axis_type=None,
x_range=[-420, 420], y_range=[-420, 420],
x_axis_label="Number of nodes (log)", y_label="distribution of degree",
min_border=0, outline_line_color=None,
background_fill="#f0e1d2", border_fill="#f0e1d2")
size = len(df)
rect(np.zeros(size), df["centers"], height=[50 for i in range(size)],
width=df["width"]*width * .8)
text(np.zeros(size), df["centers"], text=df["texts"], angle=np.zeros(size),
text_color= ["#000000"] + ["#FFFFFF" for i in xrange(size-1)],
text_align="center", text_baseline="middle")
text(np.ones(size+1) * -width/2,
[(idx-0.75)*bar_height for idx in range(size+1)],
text=map(str, map(int, bins)), angle=0)
show()
def make_trip_distance_histogram(self):
bins = self.trip_distance_bins
centers = pd.rolling_mean(bins, 2)[1:]
figure(title="trip distance in miles",
title_text_font='12pt',
plot_width=300,
plot_height=200,
x_range=[bins[0], bins[-1]],
y_range=[0, 1],
tools="pan,wheel_zoom,box_zoom,select,reset"
)
source = HistogramDataSource(
data_url="/bokeh/taxidata/distancehist/",
)
hold()
plot = rect("centers", "y", np.mean(np.diff(centers)) * 0.7, "counts",
source=source)
self.trip_distance_source = plot.select({'type' : ColumnDataSource})[0]
self.trip_distance_ar_source = source
plot.min_border=0
plot.h_symmetry=False
plot.v_symmetry=False
select_tool = _get_select_tool(plot)
if select_tool:
select_tool.dimensions = ['width']
self.distance_histogram = plot
def graph():
print("Graphing")
with open(DATA_FILE, "rb") as fin:
issues = [Issue(x) for x in pickle.load(fin)]
plotting.output_file("{}.{}.html".format(GH_USER, GH_REPO),
title="ghantt.py")
numbers = [iss.number for iss in issues]
source = ColumnDataSource(
data=dict(
number = [iss.number for iss in issues],
ago = [iss.ago + iss.length/2 for iss in issues],
length = [iss.length for iss in issues],
title = [iss.title for iss in issues],
pull_request = [iss.pull_request for iss in issues],
color = [assign_color(iss) for iss in issues],
since = ["{} days".format(int(abs(iss.ago))) for iss in issues],
),
)
plotting.hold()
plotting.rect("ago", "number", "length", 1, source=source,
color="color", title="{}/{}".format(GH_USER, GH_REPO),
y_range=(min(numbers), max(numbers)),
tools="resize,hover,previewsave,pan,wheel_zoom",
fill_alpha=0.8)
text_props = {
"source": source,
"angle": 0,
"color": "black",
"text_align": "left",
"text_baseline": "middle"
}
plotting.grid().grid_line_color = None
hover = [t for t in plotting.curplot().tools if isinstance(t, HoverTool)][0]
hover.tooltips = OrderedDict([
("number", "@number"),
("title", "@title"),
("since", "@since"),
("pull_request", "@pull_request"),
])
plotting.show()
def hist_plot(self, ticker):
global_hist, global_bins = np.histogram(self.df[ticker + "_returns"], bins=50)
hist, bins = np.histogram(self.selected_df[ticker + "_returns"], bins=50)
width = 0.7 * (bins[1] - bins[0])
center = (bins[:-1] + bins[1:]) / 2
start = global_bins.min()
end = global_bins.max()
top = hist.max()
return rect(center, hist/2.0, width, hist,
title="%s hist" % ticker,
plot_width=500, plot_height=200,
tools="",
title_text_font_size="10pt",
x_range=Range1d(start=start, end=end),
y_range=Range1d(start=0, end=top))
def hist_plot(self, ticker):
global_hist, global_bins = np.histogram(self.df[ticker + "_returns"], bins=50)
hist, bins = np.histogram(self.selected_df[ticker + "_returns"], bins=50)
width = 0.7 * (bins[1] - bins[0])
center = (bins[:-1] + bins[1:]) / 2
start = global_bins.min()
end = global_bins.max()
top = hist.max()
return rect(
center, hist/2.0, width, hist,
title="%s hist" % ticker,
plot_width=500, plot_height=200,
tools="",
title_text_font_size="10pt",
x_range=[start, end],
y_range=[0, top],
)
def drawPlot(shapeFilename, zipBorough, zipMaxAgency):
# Read the ShapeFile
dat = shapefile.Reader(shapeFilename)
# Creates a dictionary for zip: {lat_list: [], lng_list: []}.
zipCodes = []
hoverZip = []
hoverAgency = []
hoverComplaints = []
polygons = {'lat_list': [], 'lng_list': [], 'centerLat_list': [], 'centerLon_list': []}
record_index = 0
for r in dat.iterRecords():
currentZip = r[0]
# Keeps only zip codes in NY area.
if currentZip in zipMaxAgency: # was in zipBorough:
# zipCodes.append(currentZip) # moving this line into the parts loop
# Gets shape for this zip.
shape = dat.shapeRecord(record_index).shape
for part in range(len(shape.parts)):
zipCodes.append(currentZip)
hoverZip.append(currentZip)
hoverAgency.append(zipMaxAgency[currentZip][0])
hoverComplaints.append(zipMaxAgency[currentZip][1])
start = shape.parts[part]
if part == len(shape.parts) - 1:
end = len(shape.points)
else:
end = shape.parts[part + 1]
points = shape.points[start : end]
# Breaks into lists for lat/lng.
lngs = [p[0] for p in points]
lats = [p[1] for p in points]
# Calculate centers
center_lngs = min(lngs) + (max(lngs) - min(lngs))/2
center_lats = min(lats) + (max(lats) - min(lats))/2
# Store centroids for current part shape
polygons['centerLat_list'].append(center_lats)
polygons['centerLon_list'].append(center_lngs)
# Stores lat/lng for current zip shape.
polygons['lng_list'].append(lngs)
polygons['lat_list'].append(lats)
record_index += 1
# Palette
##d9d9d9
brewer11 = ['#8dd3c7', '#ffffb3', '#bebada','#fb8072','#80b1d3','#fdb462','#b3de69','#fccde5','#d9d9d9','#bc80bd','#ccebc5']
#brewer11 = ['#a6cee3', '#1f78b4','#b2df8a','#33a02c','#fb9a99','#e31a1c','#fdbf6f','#ff7f00','#cab2d6','#6a3d9a','#ffff99']
#agencies = sorted(list({zipMaxAgency[zipCode] for zipCode in zipMaxAgency}))
biggestComplaints = {}
for zz, aa in zipMaxAgency.iteritems():
if aa[0] in biggestComplaints:
biggestComplaints[aa[0]] += 1
else:
biggestComplaints[aa[0]] = 1
# sorting agencies by number of zip codes (to try to get better colors)
agencies = list(biggestComplaints.iteritems())
agencies = sorted(agencies, key = lambda x: x[1], reverse = True)
agencies = [agency[0] for agency in agencies]
# Assign colors to agencies
agencyColor = {agencies[i] : brewer11[i] for i in range(len(brewer11))}
polygons['colors'] = [agencyColor[zipMaxAgency[zipCode][0]] for zipCode in zipCodes]
# Prepare hover
#source = bk.ColumnDataSource(data=dict(hoverAgency=hoverAgency, hoverZip=hoverZip, hoverComplaintCount=hoverComplaintCount,))
source = bk.ColumnDataSource(data=dict(hoverZip = hoverZip, hoverAgency = hoverAgency, hoverComplaints = hoverComplaints),)
# Creates the Plot
bk.output_file("problem1.html")
bk.hold()
TOOLS="pan,wheel_zoom,box_zoom,reset,previewsave,hover"
fig = bk.figure(title="311 Complaints by Zip Code", \
tools=TOOLS, plot_width=800, plot_height=650)
# Creates the polygons.
bk.patches(polygons['lng_list'], polygons['lat_list'], fill_color=polygons['colors'], line_color="gray", source = source)
# RP: add hover
hover = bk.curplot().select(dict(type=HoverTool))
hover.tooltips = OrderedDict([
("Zip", "@hoverZip"),
("Agency", "@hoverAgency"),
("Number of complaints", "@hoverComplaints"),
])
### Zip codes as text on polygons
#for i in range(len(polygons['centerLat_list'])):
# y = polygons['centerLat_list'][i]
# x = polygons['centerLon_list'][i]
# zipCode = zipCodes[i]
# bk.text([x], [y], text=zipCode, angle=0, text_font_size="8pt", text_align="center", text_baseline="middle")
fonts = ["Comic sans MS", "Papyrus", "Curlz", "Impact", "Zapf dingbats", "Comic sans MS", "Papyrus", "Curlz", "Impact", "Zapf Dingbats", "Comic sans MS"]
### Legend
x = -73.66
y = 40.50
#x = -74.25
#y = 40.9
for agency, color in agencyColor.iteritems():
#print "Color: ", a
#print "x:", x
#print "y:", y
bk.rect([x], [y], color = color, width=0.03, height=0.015)
bk.text([x], [y], text = agency, angle=0, text_font_size="7pt", text_align="center", text_baseline="middle")
y = y + 0.015
#bokeh.embed.components(fig, bokeh.resources.CDN)
bk.show()
ypos = [str(y) for y in df['y']],
number = df['prime'],
color_prime = df['color']
)
)
# hold() to put all the boxes on one grid
bk.hold()
# Set x-y ranges
x_range = [str(x) for x in range(-1,11)]
y_range = [str(y) for y in range(-1,-11)]
# Make a rect glyph for each factor
bk.rect("xpos","ypos",0.9,0.9,source=source,
x_range=x_range, y_range=y_range,
fill_alpha=0.6, color="color_prime",
tools="resize", title="Prime Factor Visualization"
)
# Add text to display the number for each box
# Use text_props = dict to set properties of text elements
text_props = {
"source": source,
"angle": 0,
"color": "black",
"text_align": "center",
"text_baseline": "middle"
}
bk.text(x=dict(field="xpos", units="data"),
y=dict(field="ypos", units="data"),
def bokeh_mesh_density(meshgrid,
s_vectors=(),
s_goals=(),
swap_xy=True,
mesh_colors=('#DDDDDD', '#BBBBBB', '#999999', '#777777',
'#555555'),
title='no title',
e_radius=1.0,
e_color=E_COLOR,
e_alpha=0.75,
g_radius=1.0,
g_color=G_COLOR,
g_alpha=0.75,
x_range=None,
y_range=None,
**kwargs):
x_range, y_range = ranges(s_channels, x_range=x_range, y_range=y_range)
xm = zip(*[b.bounds[0] for b in meshgrid.nonempty_bins])
ym = zip(*[b.bounds[1] for b in meshgrid.nonempty_bins])
if swap_xy:
x_range, y_range = y_range, x_range
xm, ym = ym, xm
d_max = max(len(b) for b in meshgrid.nonempty_bins)
mesh_scale = [
1.0 + i / (len(mesh_colors) - 1) * d_max
for i in range(len(mesh_colors))
]
colorbar = ColorBar(mesh_scale, mesh_colors, continuous=True)
color = []
for b in meshgrid.nonempty_bins:
color.append(colorbar.color(len(b)))
plotting.rect(
(np.array(xm[1]) + np.array(xm[0])) / 2,
(np.array(ym[1]) + np.array(ym[0])) / 2,
(np.array(xm[1]) - np.array(xm[0])) * 0.97,
(np.array(ym[1]) - np.array(ym[0])) * 0.97,
#x_range=env.s_channels[0].bounds, y_range=env.s_channels[1].bounds,
x_range=x_range,
y_range=y_range,
fill_color=color,
fill_alpha=0.5,
line_color='#444444',
line_alpha=0.0,
title=title)
plotting.hold(True)
plotting.grid().grid_line_color = 'white'
spread(meshgrid.s_channels,
s_vectors=s_vectors,
s_goals=s_goals,
swap_xy=swap_xy,
e_radius=e_radius,
e_color=e_color,
e_alpha=e_alpha,
g_radius=g_radius,
g_color=g_color,
g_alpha=g_alpha)
plotting.hold(False)
plotting.hold(False)
for i in range(len(werte)):
anteil.append(0)
for i in range(len(anteil)):
anteil[i] = werte[i]/sumVal
print anteil
# Create `Range1d` object to set the plot's `y_range`.
yr = Range1d(start=0, end=1)
figure(y_range=yr)
# Plot each `rect` object and `show()` the plot.
hold(True)
for i in range(len(mitten)):
rect([mitten[i]],[anteil[i]/2], width=balkenbreite, height=anteil[i],
plot_width=400, color = "#ff1200", plot_height=400)
xaxis().axis_label="Areas"
yaxis().axis_label="Frequency"
# Bonus round: you can even change the `x_range` attribute of the current plot (referenced as `curplot()`) at any time. Let's give the plot a bit of padding on each side.
xr = Range1d(start=5, end=45)
curplot().x_range = xr
show()
def make_plot(xr):
yr = Range1d(start=-10, end=10)
figure(plot_width=800,
plot_height=350,
y_range=yr,
x_range=xr,
tools="xpan,xwheel_zoom,hover,box_zoom,reset")
hold()
genes = pd.read_csv('/home/hugoshi/data/lab7/genes.refseq.hg19.bed',
sep='\t',
skiprows=[0],
header=None)
genes.rename(columns={0: "chromosome", 1: "start", 2: "end"}, inplace=True)
genes = genes[genes.chromosome == 'chr5']
g_len = len(genes)
quad(
genes.start - 0.5, # left edge
genes.end - 0.5, # right edge
[2.3] * g_len, # top edge
[1.7] * g_len,
['blue'] * g_len) # bottom edge
exons = pd.read_csv('/home/hugoshi/data/lab7/exons.refseq.hg19.bed',
sep='\t',
skiprows=[0],
header=None)
exons.columns = ["chromosome", "start", "end", "meta1", "meta2", "meta3"]
exons = exons[exons.chromosome == 'chr5']
e_len = len(exons)
quad(
exons.start - 0.5, # left edge
exons.end - 0.5, # right edge
[1.3] * e_len, # top edge
[0.7] * e_len,
['blue'] * e_len) # bottom edge
df = pd.read_csv('/home/hugoshi/data/lab7/CHP2.20131001.hotspots.bed',
sep='\t',
skiprows=[0],
header=None)
df.columns = ["chromosome", "start", "end", "meta1", "meta2", "meta3"]
df = df[df.chromosome == 'chr5']
singles = df[df.start + 1 == df.end]
widers = df[df.start + 1 != df.end]
slen = len(singles)
wlen = len(widers)
s_source = ColumnDataSource(data=dict(start=singles.start,
end=singles.end,
meta1=singles.meta1,
meta2=singles.meta2,
meta3=singles.meta3))
rect(
'start', # x center
[1] * slen, # y center
[0.9] * slen,
[1] * slen,
color=['red'] * slen,
source=s_source) # height
hover = [t for t in curplot().tools if isinstance(t, HoverTool)][0]
hover.tooltips = OrderedDict([
# add to this
("position", "@start"),
("meta 1", "@meta1"),
("meta 2", "@meta2"),
("meta 3", "@meta3")
])
quad(
widers.start - 0.5, # left edge
widers.end - 0.5, # right edge
[0.3] * wlen, # top edge
[-0.3] * wlen) # bottom edge
hold()
return curplot()
name = df['cabinetName'],
loc = df['labLocation'],
sub = df['subsystem'],
equip = df['equipment'],
phone = df['phone'],
color_type = [colormap[str(x)] for x in df['subsystem']]
)
)
# hold() to put all glyphs on the same graph
bk.hold()
# Make a rect glyph for each element of data
bk.rect("posx", "posy", 0.9, 0.9, source=source,
x_range=x_range, y_range=y_range,
line_color="color_type", line_cap='round',
fill_alpha=0.6, color="color_type",
tools="resize,hover,previewsave,ywheel_zoom", title="ESS Lab Layout",
plot_width=1600, plot_height=3200
)
# Use text_props = dict to set properties of text elements
text_props = {
"source": source,
"angle": 0,
"color": "black",
"text_align": "center",
"text_baseline": "middle"
}
# Make a text glyph for each rect glyph to show the cabinet names and locations
bk.text(x=dict(field="posx", units="data"),
def bokeh_mesh_density(
meshgrid,
s_vectors=(),
s_goals=(),
swap_xy=True,
mesh_colors=("#DDDDDD", "#BBBBBB", "#999999", "#777777", "#555555"),
title="no title",
e_radius=1.0,
e_color=E_COLOR,
e_alpha=0.75,
g_radius=1.0,
g_color=G_COLOR,
g_alpha=0.75,
x_range=None,
y_range=None,
**kwargs
):
x_range, y_range = ranges(s_channels, x_range=x_range, y_range=y_range)
xm = zip(*[b.bounds[0] for b in meshgrid.nonempty_bins])
ym = zip(*[b.bounds[1] for b in meshgrid.nonempty_bins])
if swap_xy:
x_range, y_range = y_range, x_range
xm, ym = ym, xm
d_max = max(len(b) for b in meshgrid.nonempty_bins)
mesh_scale = [1.0 + i / (len(mesh_colors) - 1) * d_max for i in range(len(mesh_colors))]
colorbar = ColorBar(mesh_scale, mesh_colors, continuous=True)
color = []
for b in meshgrid.nonempty_bins:
color.append(colorbar.color(len(b)))
plotting.rect(
(np.array(xm[1]) + np.array(xm[0])) / 2,
(np.array(ym[1]) + np.array(ym[0])) / 2,
(np.array(xm[1]) - np.array(xm[0])) * 0.97,
(np.array(ym[1]) - np.array(ym[0])) * 0.97,
# x_range=env.s_channels[0].bounds, y_range=env.s_channels[1].bounds,
x_range=x_range,
y_range=y_range,
fill_color=color,
fill_alpha=0.5,
line_color="#444444",
line_alpha=0.0,
title=title,
)
plotting.hold(True)
plotting.grid().grid_line_color = "white"
spread(
meshgrid.s_channels,
s_vectors=s_vectors,
s_goals=s_goals,
swap_xy=swap_xy,
e_radius=e_radius,
e_color=e_color,
e_alpha=e_alpha,
g_radius=g_radius,
g_color=g_color,
g_alpha=g_alpha,
)
plotting.hold(False)
plotting.hold(False)
def drawPlot(shapeFilename, zipBorough, zipMaxAgency):
# Read the ShapeFile
dat = shapefile.Reader(shapeFilename)
# Creates a dictionary for zip: {lat_list: [], lng_list: []}.
zipCodes = []
hoverZip = []
hoverAgency = []
hoverComplaints = []
polygons = {
'lat_list': [],
'lng_list': [],
'centerLat_list': [],
'centerLon_list': []
}
record_index = 0
for r in dat.iterRecords():
currentZip = r[0]
# Keeps only zip codes in NY area.
if currentZip in zipMaxAgency: # was in zipBorough:
# zipCodes.append(currentZip) # moving this line into the parts loop
# Gets shape for this zip.
shape = dat.shapeRecord(record_index).shape
for part in range(len(shape.parts)):
zipCodes.append(currentZip)
hoverZip.append(currentZip)
hoverAgency.append(zipMaxAgency[currentZip][0])
hoverComplaints.append(zipMaxAgency[currentZip][1])
start = shape.parts[part]
if part == len(shape.parts) - 1:
end = len(shape.points)
else:
end = shape.parts[part + 1]
points = shape.points[start:end]
# Breaks into lists for lat/lng.
lngs = [p[0] for p in points]
lats = [p[1] for p in points]
# Calculate centers
center_lngs = min(lngs) + (max(lngs) - min(lngs)) / 2
center_lats = min(lats) + (max(lats) - min(lats)) / 2
# Store centroids for current part shape
polygons['centerLat_list'].append(center_lats)
polygons['centerLon_list'].append(center_lngs)
# Stores lat/lng for current zip shape.
polygons['lng_list'].append(lngs)
polygons['lat_list'].append(lats)
record_index += 1
# Palette
##d9d9d9
brewer11 = [
'#8dd3c7', '#ffffb3', '#bebada', '#fb8072', '#80b1d3', '#fdb462',
'#b3de69', '#fccde5', '#d9d9d9', '#bc80bd', '#ccebc5'
]
#brewer11 = ['#a6cee3', '#1f78b4','#b2df8a','#33a02c','#fb9a99','#e31a1c','#fdbf6f','#ff7f00','#cab2d6','#6a3d9a','#ffff99']
#agencies = sorted(list({zipMaxAgency[zipCode] for zipCode in zipMaxAgency}))
biggestComplaints = {}
for zz, aa in zipMaxAgency.iteritems():
if aa[0] in biggestComplaints:
biggestComplaints[aa[0]] += 1
else:
biggestComplaints[aa[0]] = 1
# sorting agencies by number of zip codes (to try to get better colors)
agencies = list(biggestComplaints.iteritems())
agencies = sorted(agencies, key=lambda x: x[1], reverse=True)
agencies = [agency[0] for agency in agencies]
# Assign colors to agencies
agencyColor = {agencies[i]: brewer11[i] for i in range(len(brewer11))}
polygons['colors'] = [
agencyColor[zipMaxAgency[zipCode][0]] for zipCode in zipCodes
]
# Prepare hover
#source = bk.ColumnDataSource(data=dict(hoverAgency=hoverAgency, hoverZip=hoverZip, hoverComplaintCount=hoverComplaintCount,))
source = bk.ColumnDataSource(data=dict(hoverZip=hoverZip,
hoverAgency=hoverAgency,
hoverComplaints=hoverComplaints), )
# Creates the Plot
bk.output_file("problem1.html")
bk.hold()
TOOLS = "pan,wheel_zoom,box_zoom,reset,previewsave,hover"
fig = bk.figure(title="311 Complaints by Zip Code", \
tools=TOOLS, plot_width=800, plot_height=650)
# Creates the polygons.
bk.patches(polygons['lng_list'],
polygons['lat_list'],
fill_color=polygons['colors'],
line_color="gray",
source=source)
# RP: add hover
hover = bk.curplot().select(dict(type=HoverTool))
hover.tooltips = OrderedDict([
("Zip", "@hoverZip"),
("Agency", "@hoverAgency"),
("Number of complaints", "@hoverComplaints"),
])
### Zip codes as text on polygons
#for i in range(len(polygons['centerLat_list'])):
# y = polygons['centerLat_list'][i]
# x = polygons['centerLon_list'][i]
# zipCode = zipCodes[i]
# bk.text([x], [y], text=zipCode, angle=0, text_font_size="8pt", text_align="center", text_baseline="middle")
fonts = [
"Comic sans MS", "Papyrus", "Curlz", "Impact", "Zapf dingbats",
"Comic sans MS", "Papyrus", "Curlz", "Impact", "Zapf Dingbats",
"Comic sans MS"
]
### Legend
x = -73.66
y = 40.50
#x = -74.25
#y = 40.9
for agency, color in agencyColor.iteritems():
#print "Color: ", a
#print "x:", x
#print "y:", y
bk.rect([x], [y], color=color, width=0.03, height=0.015)
bk.text([x], [y],
text=agency,
angle=0,
text_font_size="7pt",
text_align="center",
text_baseline="middle")
y = y + 0.015
#bokeh.embed.components(fig, bokeh.resources.CDN)
bk.show()
def make_plot(xr):
yr = Range1d(start=-10, end=10)
figure(plot_width=800, plot_height=350,
y_range=yr,
x_range=xr,
tools="xpan,xwheel_zoom,hover,box_zoom,reset")
hold()
genes = pd.read_csv('/home/hugoshi/data/lab7/genes.refseq.hg19.bed', sep='\t',
skiprows=[0], header=None)
genes.rename(columns={
0: "chromosome",
1: "start",
2: "end"},
inplace=True
)
genes = genes[genes.chromosome == 'chr5']
g_len = len(genes)
quad(genes.start - 0.5, # left edge
genes.end - 0.5, # right edge
[2.3] * g_len, # top edge
[1.7] * g_len,
['blue'] * g_len) # bottom edge
exons = pd.read_csv('/home/hugoshi/data/lab7/exons.refseq.hg19.bed', sep='\t', skiprows=[0], header=None)
exons.columns = ["chromosome", "start", "end", "meta1", "meta2", "meta3"]
exons = exons[exons.chromosome == 'chr5']
e_len = len(exons)
quad(exons.start - 0.5, # left edge
exons.end - 0.5, # right edge
[1.3] * e_len, # top edge
[0.7] * e_len,
['blue'] * e_len) # bottom edge
df = pd.read_csv('/home/hugoshi/data/lab7/CHP2.20131001.hotspots.bed', sep='\t', skiprows=[0], header=None)
df.columns = ["chromosome", "start", "end", "meta1", "meta2", "meta3"]
df = df[df.chromosome == 'chr5']
singles = df[df.start+1 == df.end]
widers = df[df.start+1 != df.end]
slen = len(singles)
wlen = len(widers)
s_source = ColumnDataSource(
data = dict(
start=singles.start,
end=singles.end,
meta1=singles.meta1,
meta2=singles.meta2,
meta3=singles.meta3))
rect('start', # x center
[1]*slen, # y center
[0.9]*slen,
[1]*slen,
color=['red']*slen,
source=s_source) # height
hover = [t for t in curplot().tools if isinstance(t, HoverTool)][0]
hover.tooltips = OrderedDict([
# add to this
("position", "@start"),
("meta 1", "@meta1"),
("meta 2", "@meta2"),
("meta 3", "@meta3")
])
quad(widers.start - 0.5, # left edge
widers.end - 0.5, # right edge
[0.3] * wlen, # top edge
[-0.3] * wlen) # bottom edge
hold()
return curplot()
def make_box_violin_plot_2(data, maxwidth=0.9):
"""
data: dict[Str -> List[Number]]
maxwidth: float
Maximum width of tornado plot within each factor/facet
Returns the plot object
"""
df = pd.DataFrame(columns=["group", "centers", "width", "height", "texts"])
bar_height = 50
bins = [0, 1e1, 1e2, 1e3, 1e4, 1e5, 1e6]
# Compute histograms, while keeping track of max Y values and max counts
for i, (group, vals) in enumerate(data.iteritems()):
hist, edges = np.histogram(vals, bins)
df = df.append(
pd.DataFrame(
dict(
group=group,
centers=np.arange(len(hist)) * bar_height,
width=np.log10(hist),
height=np.ones(hist.shape) * bar_height,
texts=map(str, hist),
)))
df.replace(-np.inf, 0)
# Normalize the widths
df["width"] *= (maxwidth / df["width"].max())
hold()
width, height = 800, 800
figure(plot_width=width,
plot_height=height,
title="Degree Distribution",
tools="previewsave",
x_axis_type=None,
y_axis_type=None,
x_range=[-420, 420],
y_range=[-420, 420],
x_axis_label="Number of nodes (log)",
y_label="distribution of degree",
min_border=0,
outline_line_color=None,
background_fill="#f0e1d2",
border_fill="#f0e1d2")
size = len(df)
rect(np.zeros(size),
df["centers"],
height=[50 for i in range(size)],
width=df["width"] * width * .8)
text(np.zeros(size),
df["centers"],
text=df["texts"],
angle=np.zeros(size),
text_color=["#000000"] + ["#FFFFFF" for i in xrange(size - 1)],
text_align="center",
text_baseline="middle")
text(np.ones(size + 1) * -width / 2,
[(idx - 0.75) * bar_height for idx in range(size + 1)],
text=map(str, map(int, bins)),
angle=0)
show()
