def bokeh_nn(
s_channels, testset, errors, title="no title", swap_xy=True, x_range=None, y_range=None, radius=3.0, alpha=0.75
):
x_range, y_range = ranges(s_channels, x_range=x_range, y_range=y_range)
xv, yv = zip(*testset)
if swap_xy:
x_range, y_range = y_range, x_range
xv, yv = yv, xv
scale = [0, max(errors)]
colorbar = ColorBar(scale, ["#0000FF", "#FF0000"], continuous=True)
colors = [colorbar.color(e) for e in errors]
plotting.scatter(
xv,
yv,
title=title,
x_range=x_range,
y_range=y_range,
fill_color=colors,
fill_alpha=alpha,
line_color=None,
radius=radius,
radius_units="screen",
)
plotting.hold(True)
plotting.grid().grid_line_color = "white"
plotting.hold(False)
def bokeh_nn(s_channels,
testset,
errors,
title='no title',
swap_xy=True,
x_range=None,
y_range=None,
radius=3.0,
alpha=0.75):
x_range, y_range = ranges(s_channels, x_range=x_range, y_range=y_range)
xv, yv = zip(*testset)
if swap_xy:
x_range, y_range = y_range, x_range
xv, yv = yv, xv
scale = [0, max(errors)]
colorbar = ColorBar(scale, ['#0000FF', '#FF0000'], continuous=True)
colors = [colorbar.color(e) for e in errors]
plotting.scatter(xv,
yv,
title=title,
x_range=x_range,
y_range=y_range,
fill_color=colors,
fill_alpha=alpha,
line_color=None,
radius=radius,
radius_units="screen")
plotting.hold(True)
plotting.grid().grid_line_color = 'white'
plotting.hold(False)
def plotScatter(data):
data_to_plot = data.dropna()
pl.output_file("scatter.html")
pl.figure()
pl.hold()
pl.scatter(data["Fare"],data["Age"],
color=data["Survived"].map(colormap),alpha=0.8,
xlabel="Fare")
pl.show()
def bokeh_plot(data, outfile):
table = np.genfromtxt(data, names=['x', 'y'])
blt.output_file(outfile)
blt.hold()
blt.figure(tools="pan,wheel_zoom,box_zoom,reset,previewsave,select")
blt.scatter(table['x'], table['y'])
blt.show()
def build_plot(datalist,logx=True):
# Set the output for our plot.
output_file('plot.html', title='Plot')#, js="relative", css="relative")
# Create some data for our plot.
colors=['red','green','blue','yellow','black']
# colors=['tomato','navy']
hold()
cnt=0
for name,data in datalist:
cnt+=1
x,y=zip(*data)
if logx:
scatter([np.log(x_el) for x_el in x] ,list(y) , color=colors[cnt], legend=name)
else:
scatter(list(x) ,list(y) , color=colors[cnt], legend=name)
# Create an HTML snippet of our plot.
snippet = curplot().create_html_snippet(embed_base_url='/static/plots/', embed_save_loc=plotdir)
# Return the snippet we want to place in our page.
return snippet
def example_hyperlink_plot():
idx_df = pds.read_csv("example_index.txt", sep="\t", names=["url", "id"])
arc_df = pds.read_csv("example_arcs.txt", sep="\t", names=["src", "dest"])
out_deg = arc_df.groupby(["src"]).size()
out_deg.sort(ascending=False)
in_deg = arc_df.groupby(["dest"]).size()
in_deg.sort(ascending=False)
color_map_max = out_deg.max()
pallette = b2m.get_map("Dark2", "Qualitative", 8).hex_colors
num_colors = len(pallette) - 1
color_map = lambda x: pallette[int(num_colors*(x/color_map_max))]
color_vals = out_deg[in_deg.index].fillna(0)
colors = color_vals.map(color_map)
plt.output_file("in_deg.html", title="In degree rankings")
plt.scatter(range(in_deg.size), in_deg, color=colors, fill_alpha=0.2,
size=10, name="In Degree")
plt.show()
def xy_scatter_bokeh(self):
from bokeh.plotting import output_server, figure, scatter, show
ux, uy, pos_mean, pos_sem, neg_mean, neg_sem, colors, size = \
self.xy_plot_info()
output_server("scatter.html")
figure(tools="pan,wheel_zoom,box_zoom,reset,previewsave,select")
scatter(ux, uy, color="#FF00FF", nonselection_fill_color="#FFFF00", nonselection_fill_alpha=1)
scatter(ux, uy, color="red")
scatter(ux, uy, marker="square", color="green")
scatter(ux, uy, marker="square", color="blue", name="scatter_example")
show()
def plot_summary(normalized_counts, chromosome):
#print " - plotting " + chromosome + " summary"
condition = "(file_name == '*') & (chromosome == '%s')" % chromosome
chromosome_count_by_bin = collections.defaultdict(int)
for row in normalized_counts.where(condition):
chromosome_count_by_bin[row["bin_number"]] += row["count"]
num_bins = len(chromosome_count_by_bin)
if num_bins < 2:
print "-- skipping plotting %s because not enough bins (only %d)" % (chromosome, num_bins)
return
overall = bp.scatter(chromosome_count_by_bin.keys(), chromosome_count_by_bin.values())
overall.title = chromosome + " counts per bin across all bam files"
def plot_summary(normalized_counts, chromosome):
logging.debug(" - plotting %s summary", chromosome)
condition = "(file_key == 0) & (chromosome == '%s')" % chromosome
chromosome_count_by_bin = collections.defaultdict(int)
for row in normalized_counts.where(condition):
chromosome_count_by_bin[row["bin_number"]] += row["count"]
num_bins = len(chromosome_count_by_bin)
if num_bins < 2:
logging.info(
"-- skipping plotting %s because not enough bins (only %d)",
chromosome, num_bins)
return
overall = bp.scatter(chromosome_count_by_bin.keys(),
chromosome_count_by_bin.values())
overall.title = chromosome + " counts per bin across all bam files"
def plot(output_directory, normalized_counts, chromosome, cell_types):
bp.output_file(output_directory + "/" + chromosome + ".html")
plot_summary(normalized_counts, chromosome)
for cell_type in cell_types:
#print " - plotting " + cell_type
bin_number = []
count = []
condition = "(file_name == '*') & (chromosome == '%s') & (cell_type == '%s')" % (chromosome, cell_type)
for row in normalized_counts.where(condition):
bin_number.append(row["bin_number"])
count.append(row["count"])
cell_type_plot = bp.scatter(bin_number, count)
cell_type_plot.title = "%s counts per bin" % cell_type
bp.save()
def plot(output_directory, normalized_counts, chromosome, cell_types):
bp.output_file(output_directory + "/" + chromosome + ".html")
plot_summary(normalized_counts, chromosome)
for cell_type in cell_types:
logging.debug(" - plotting %s", cell_type)
bin_number = []
count = []
condition = "(file_key == 0) & (chromosome == '%s') & (cell_type == '%s')" % (
chromosome, cell_type)
for row in normalized_counts.where(condition):
bin_number.append(row["bin_number"])
count.append(row["count"])
cell_type_plot = bp.scatter(bin_number, count)
cell_type_plot.title = "%s counts per bin" % cell_type
bp.save()
def plot_summary(chromosome, common_clause):
print "Plotting " + chromosome + " summary"
cursor = db.cursor()
overall_sql = ("SELECT bin, SUM(count_fraction) AS count FROM normalized_bins WHERE " +
common_clause + " GROUP BY bin")
cursor.execute(overall_sql)
# todo: I should probably be able to do something like the following without copying:
# bp.scatter(cursor.fetchall())
bin_number = []
count = []
for row in cursor.fetchall():
bin_number.append(int(row[0]))
count.append(float(row[1]))
overall = bp.scatter(bin_number, count)
overall.title = chromosome + " counts per bin across all bam files"
def create_scatter_plot(releases):
number_of_bidders = []
delta_time = []
for release in releases:
start = get_start_date(release)
award = get_award_date(release)
delta = get_difference_in_days(start, award)
delta_time.append(delta)
number_of_bidders.append(get_number_of_bidders(release))
scatter_plot = scatter(delta_time, number_of_bidders,
fill_alpha=0.2, size=10,
name="period between start of tender period and award date vs amount of bidders")
scatter_plot.title = "period vs. bidders"
xaxis()[0].axis_label = "Days"
yaxis()[0].axis_label = "Number of Bidders"
show()
def plot(chromosome, common_clause, cell_types):
bp.output_file(chromosome + ".html")
plot_summary(chromosome, common_clause)
for cell_type in cell_types:
print "Plotting " + cell_type
bin_number = []
count = []
cell_type_bin_sql = ("SELECT bin, count_fraction FROM normalized_bins WHERE " +
common_clause + "AND cell_type = '%s'" % cell_type)
cell_type_cursor = db.cursor()
cell_type_cursor.execute(cell_type_bin_sql)
for cell_type_row in cell_type_cursor.fetchall():
bin_number.append(int(cell_type_row[0]))
count.append(float(cell_type_row[1]))
cell_type_plot = bp.scatter(bin_number, count)
cell_type_plot.title = "%s counts per bin" % cell_type
bp.save()
def spread(
s_channels,
s_vectors=(),
s_goals=(),
title="no title",
swap_xy=True,
x_range=None,
y_range=None,
e_radius=1.0,
e_color=E_COLOR,
e_alpha=0.75,
g_radius=1.0,
g_color=G_COLOR,
g_alpha=0.75,
grid=True,
radius_units="screen",
font_size="11pt",
**kwargs
):
x_range, y_range = ranges(s_channels, x_range=x_range, y_range=y_range)
# effects
try:
xv, yv = zip(*(s[:2] for s in s_vectors))
except ValueError:
xv, yv = [], []
if swap_xy:
x_range, y_range = y_range, x_range
xv, yv = yv, xv
plotting.scatter(
xv,
yv,
title=title,
x_range=x_range,
y_range=y_range,
fill_color=e_color,
fill_alpha=e_alpha,
line_color=None,
radius=e_radius,
radius_units=radius_units,
**kwargs
)
# title_text_font_size=font_size, **kwargs)
plotting.hold(True)
# goals
try:
xg, yg = zip(*s_goals)
except ValueError:
xg, yg = [], []
if swap_xy:
xg, yg = yg, xg
plotting.scatter(
xg, yg, radius=g_radius, radius_units="screen", fill_color=g_color, fill_alpha=g_alpha, line_color=None
)
plotting_axis()
if not grid:
plotting.grid().grid_line_color = None
plotting.hold(False)
def plot1(n=200):
plotting.figure(title='arm1, {} steps'.format(n))
plotting.scatter(arm1_xs[:n], arm1_ys[:n],
x_range=[-1, 1], y_range=[-1, 1],
fill_alpha= 0.5, line_color=None, radius=2.0, radius_units='screen')
def drawPlot(shapeFilename, zipBorough, grids, centers, gridLines):
# Read the ShapeFile
dat = shapefile.Reader(shapeFilename)
# Creates a dictionary for zip: {lat_list: [], lng_list: []}.
zipCodes = []
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 zipBorough: # was in zipBorough:
# Gets shape for this zip.
shape = dat.shapeRecord(record_index).shape
for part in range(len(shape.parts)):
zipCodes.append(currentZip)
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
# Creates the Plot
bk.output_file("problem3.html")
bk.hold()
north = 40.915
east = -73.651
south = 40.496
west = -74.256
width = east - west
height = north - south
x_range = [west - 0.05 * width, east + 0.05 * width]
y_range = [south - 0.05 * height, north + 0.05 * height]
TOOLS="pan,wheel_zoom,box_zoom,reset,previewsave"
fig = bk.figure(title="311 Complaints by Zip Code", \
tools=TOOLS, background_fill = "#f8f8f8", x_range = x_range, y_range = y_range)
circleSizes = [item for sublist in grids['log_counts'] for item in sublist]
# Creates the polygons
cellWidth = gridLines['vLines'][1]-gridLines['vLines'][0]
cellHeight = gridLines['hLines'][1] - gridLines['hLines'][0]
bk.patches(polygons['lng_list'], polygons['lat_list'], fill_color="#fee8c8", line_color="gray")
circleSizes = [0.009 * (size ** 3.7) for size in circleSizes]
bk.scatter(centers['lon'], centers['lat'], size = circleSizes, alpha = 0.4, line_color = None, fill_color = 'red')
bk.hold()
#print type(centers['lon'])
circleSizeArr = np.array(circleSizes)
maxSize = np.max(circleSizeArr)
circleSizeArr[circleSizeArr == 0] = 100
minSize = np.min(circleSizeArr)
#print minSize, maxSize
legendN = 5
legendCircles = list(np.linspace(minSize, maxSize, legendN))
legendCounts = [str(int(math.exp(((size/0.009) ** (1.0/3.7))))) + " complaints" for size in legendCircles]
#print legendCircles
fakeScatter = [0 for x in range(legendN)]
fakeScatterX = [-74.23 for x in range(legendN)]
fakeScatterXtext = [-74.23 + 0.05 for x in range(legendN)]
#fakeScatterY = [40.8 + 0.028*y for y in range(legendN)]
fakeScatterY = [40.8, 40.8 + 0.02, 40.8 + 0.036, 40.8 + 0.056,40.8 + 0.083]
#print fakeScatterX, type(fakeScatterX)
#print fakeScatterY, type(fakeScatterY)
bk.scatter(fakeScatterX, fakeScatterY, size = legendCircles, fill_color = 'red', line_color = None, alpha = 0.4)
bk.text(fakeScatterXtext[1:], fakeScatterY[1:], text = legendCounts[1:], angle = 0, text_align = "left", text_font_size = "7pt", text_baseline = 'middle')
#for i in range(len(fakeScatter)):
# bk.scatter([fakeScatter[i]], [fakeScatter[i]], size = [legendCircles[i]], legend = int(legendCircles[i]), color = 'red', fill_line = None, alpha = 0.4)
# Disable background grid
bk.grid().grid_line_color = None
#print centers
### Legend
#x = -74.25
#y = 40.8
#height = 0.003
#legend_box_y = y + 0.5 * height * len(legendColors)
#bk.rect([x+0.032], [legend_box_y], width = 0.12, height = height*len(legendColors)*1.15, color = "#ffffff", line_color = "gray")
##x = -74.25
##y = 40.9
#for color in legendColors:
## #print "Color: ", a
## #print "x:", x
## #print "y:", y
##
# bk.rect([x], [y], color = color, width=0.03, height=height)
# #bk.text([x], [y], text = agency, angle=0, text_font_size="7pt", text_align="center", text_baseline="middle")
# y = y + height
#legend_bottom_y = 40.797
#legend_top_y = legend_bottom_y + 0.92 * height * len(legendColors)
#bk.text([-74.225], [legend_bottom_y], text = agency2, angle = 0, text_font_size = "7pt", text_align = "left")
#bk.text([-74.225], [legend_top_y], text = agency1, angle = 0, text_font_size = "7pt", text_align = "left")
#bokeh.embed.components(fig, bokeh.resources.CDN)
bk.show()
# Running the Target Exploration
tgt_explorations = []
for i in range(N):
exploration = ex.explore()
feedback = env.execute(exploration['m_signal'])
tgt_ex.receive(exploration, feedback)
tgt_explorations.append((exploration, feedback))
tgt_dataset = {'m_channels' : ex.m_channels,
's_channels' : ex.s_channels,
'explorations': tgt_explorations}
# Graphs
try:
from bokeh import plotting
plotting.output_file('goal_reuse.html')
xs = [explo[1]['s_signal']['x'] for explo in dataset['explorations']]
ys = [explo[1]['s_signal']['y'] for explo in dataset['explorations']]
plotting.scatter(xs, ys, x_range=[-1, 1], y_range=[-1, 1], title='explorers library: goal reuse algorithm example',
fill_alpha= 0.5, line_color=None, radius=2.0, radius_units='screen')
plotting.show()
except ImportError:
print('exploration went fine, but you need the bokeh library (http://bokeh.pydata.org/) to display it')
feedback = env.execute(exploration['m_signal'])
ex.receive(exploration, feedback)
explorations.append((exploration, feedback))
dataset = {'m_channels' : ex.m_channels,
's_channels' : ex.s_channels,
'explorations': explorations}
# Graph
# the red dots are from motor babbling
# the blue dots from goal babbling
try:
from bokeh import plotting
plotting.output_file('goal_babbling.html')
xs = [explo[1]['s_signal']['x'] for explo in dataset['explorations']]
ys = [explo[1]['s_signal']['y'] for explo in dataset['explorations']]
plotting.scatter(xs, ys, x_range=[-1, 1], y_range=[-1, 1], title='explorers library: exploration example',
fill_alpha= 0.5, line_color=None, radius=2.0, radius_units='screen')
plotting.hold(True)
plotting.scatter(xs[:ex_cfg.eras[0]], ys[:ex_cfg.eras[0]], x_range=[-1, 1], y_range=[-1, 1], title='explorers library: exploration example',
fill_alpha= 0.5, color='red', line_color=None, radius=2.0, radius_units='screen')
plotting.show()
except ImportError:
print('exploration went fine, but you need the bokeh library to display the it')
def spread(s_channels,
s_vectors=(),
s_goals=(),
title='no title',
swap_xy=True,
x_range=None,
y_range=None,
e_radius=1.0,
e_color=E_COLOR,
e_alpha=0.75,
g_radius=1.0,
g_color=G_COLOR,
g_alpha=0.75,
grid=True,
radius_units='screen',
font_size='11pt',
**kwargs):
x_range, y_range = ranges(s_channels, x_range=x_range, y_range=y_range)
# effects
try:
xv, yv = zip(*(s[:2] for s in s_vectors))
except ValueError:
xv, yv = [], []
if swap_xy:
x_range, y_range = y_range, x_range
xv, yv = yv, xv
plotting.scatter(xv,
yv,
title=title,
x_range=x_range,
y_range=y_range,
fill_color=e_color,
fill_alpha=e_alpha,
line_color=None,
radius=e_radius,
radius_units=radius_units,
**kwargs)
# title_text_font_size=font_size, **kwargs)
plotting.hold(True)
# goals
try:
xg, yg = zip(*s_goals)
except ValueError:
xg, yg = [], []
if swap_xy:
xg, yg = yg, xg
plotting.scatter(xg,
yg,
radius=g_radius,
radius_units="screen",
fill_color=g_color,
fill_alpha=g_alpha,
line_color=None)
plotting_axis()
if not grid:
plotting.grid().grid_line_color = None
plotting.hold(False)
def cluster_vals(nodes, values):
plt.output_file("in_deg.html", title="In degree rankings")
plt.scatter(nodes, values, size=10, name="Cluster Values")
plt.show()
def drawPlot(shapeFilename, zipBorough, zipCount):
# Read the ShapeFile
dat = shapefile.Reader(shapeFilename)
# Creates a dictionary for zip: {lat_list: [], lng_list: []}.
zipCodes = []
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 zipBorough: # was in zipBorough:
# Gets shape for this zip.
shape = dat.shapeRecord(record_index).shape
### Only first polygon if non-contiguous zip codes
#firstPoint = shape.parts[0]
#if len(shape.parts) > 1:
# lastPoint = shape.parts[1]
#else:
# lastPoint = len(shape.points)
#zipCodes.append(currentZip)
#points = shape.points[firstPoint : lastPoint]
#
## 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)
### All shapes for each zip code
for part in range(len(shape.parts)):
zipCodes.append(currentZip)
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
# Creates the Plot
bk.output_file("problem4.html")
bk.hold()
north = 40.915
east = -73.651
south = 40.496
west = -74.256
width = east - west
height = north - south
x_range = [west - 0.05 * width, east + 0.05 * width]
y_range = [south - 0.05 * height, north + 0.05 * height]
TOOLS="pan,wheel_zoom,box_zoom,reset,previewsave"
fig = bk.figure(title="311 Complaints by Zip Code", \
tools=TOOLS, background_fill = "#f8f8f8", x_range = x_range, y_range = y_range, plot_height=800, plot_width = 1200)
#circleSizes = [zipCount[zipCode] for zipCode in zipCodes]
circleSizes = []
for zipCode in zipCodes:
if zipCode in zipCount:
circleSizes.append(zipCount[zipCode])
else:
circleSizes.append(1) # for taking logs
#print circleSizes
#print "aaaaaaaaaaaaaaaaaaaaa: ", len(polygons['lng_list']), len(circleSizes)
logCircleSizes = ((np.log(np.array(circleSizes)))**3.7) * 0.009
#print logCircleSizes
logCircleSizes = list(logCircleSizes)
#circleSizes = [x for x in range(len(zipCodes))]
# Creates the polygons
bk.patches(polygons['lng_list'], polygons['lat_list'], fill_color="#fee8c8", line_color="gray")
bk.scatter(polygons['centerLon_list'], polygons['centerLat_list'], size = logCircleSizes, alpha = 0.4, line_color = None, fill_color = 'red')
#circleSizes = [0.009 * (size ** 3.7) for size in circleSizes]
#bk.scatter(centers['lon'], centers['lat'], size = circleSizes, alpha = 0.4, line_color = None, fill_color = 'red')
#circleSizeArr = np.array(circleSizes)
#maxSize = np.max(circleSizeArr)
#circleSizeArr[circleSizeArr == 0] = 100
#minSize = np.min(circleSizeArr)
#legendN = 5
#legendCircles = list(np.linspace(minSize, maxSize, legendN))
#legendCounts = [str(int((size ** (1.0/3.7))/0.009)) + " complaints" for size in legendCircles]
#fakeScatter = [0 for x in range(legendN)]
#fakeScatterX = [-74.23 for x in range(legendN)]
#fakeScatterXtext = [-74.23 + 0.05 for x in range(legendN)]
#fakeScatterY = [40.8, 40.8 + 0.02, 40.8 + 0.036, 40.8 + 0.056,40.8 + 0.083]
#bk.scatter(fakeScatterX, fakeScatterY, size = legendCircles, fill_color = 'red', line_color = None, alpha = 0.4)
#bk.text(fakeScatterXtext[1:], fakeScatterY[1:], text = legendCounts[1:], angle = 0, text_align = "left", text_font_size = "7pt", text_baseline = 'middle')
# Disable background grid
bk.grid().grid_line_color = None
bk.show()
def run():
colormap = {'setosa': 'red', 'versicolor': 'green', 'virginica': 'blue'}
flowers['color'] = flowers['species'].map(lambda x: colormap[x])
return scatter(flowers["petal_length"], flowers["petal_width"], color=flowers["color"], fill_alpha=0.2, size=10, name="iris")
def drawPlot(shapeFilename, zipBorough, grids, centers, gridLines):
# Read the ShapeFile
dat = shapefile.Reader(shapeFilename)
# Creates a dictionary for zip: {lat_list: [], lng_list: []}.
zipCodes = []
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 zipBorough: # was in zipBorough:
# Gets shape for this zip.
shape = dat.shapeRecord(record_index).shape
for part in range(len(shape.parts)):
zipCodes.append(currentZip)
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
# Creates the Plot
bk.output_file("problem3.html")
bk.hold()
north = 40.915
east = -73.651
south = 40.496
west = -74.256
width = east - west
height = north - south
x_range = [west - 0.05 * width, east + 0.05 * width]
y_range = [south - 0.05 * height, north + 0.05 * height]
TOOLS = "pan,wheel_zoom,box_zoom,reset,previewsave"
fig = bk.figure(title="311 Complaints by Zip Code", \
tools=TOOLS, background_fill = "#f8f8f8", x_range = x_range, y_range = y_range)
circleSizes = [item for sublist in grids['log_counts'] for item in sublist]
# Creates the polygons
cellWidth = gridLines['vLines'][1] - gridLines['vLines'][0]
cellHeight = gridLines['hLines'][1] - gridLines['hLines'][0]
bk.patches(polygons['lng_list'],
polygons['lat_list'],
fill_color="#fee8c8",
line_color="gray")
circleSizes = [0.009 * (size**3.7) for size in circleSizes]
bk.scatter(centers['lon'],
centers['lat'],
size=circleSizes,
alpha=0.4,
line_color=None,
fill_color='red')
bk.hold()
#print type(centers['lon'])
circleSizeArr = np.array(circleSizes)
maxSize = np.max(circleSizeArr)
circleSizeArr[circleSizeArr == 0] = 100
minSize = np.min(circleSizeArr)
#print minSize, maxSize
legendN = 5
legendCircles = list(np.linspace(minSize, maxSize, legendN))
legendCounts = [
str(int(math.exp(((size / 0.009)**(1.0 / 3.7))))) + " complaints"
for size in legendCircles
]
#print legendCircles
fakeScatter = [0 for x in range(legendN)]
fakeScatterX = [-74.23 for x in range(legendN)]
fakeScatterXtext = [-74.23 + 0.05 for x in range(legendN)]
#fakeScatterY = [40.8 + 0.028*y for y in range(legendN)]
fakeScatterY = [
40.8, 40.8 + 0.02, 40.8 + 0.036, 40.8 + 0.056, 40.8 + 0.083
]
#print fakeScatterX, type(fakeScatterX)
#print fakeScatterY, type(fakeScatterY)
bk.scatter(fakeScatterX,
fakeScatterY,
size=legendCircles,
fill_color='red',
line_color=None,
alpha=0.4)
bk.text(fakeScatterXtext[1:],
fakeScatterY[1:],
text=legendCounts[1:],
angle=0,
text_align="left",
text_font_size="7pt",
text_baseline='middle')
#for i in range(len(fakeScatter)):
# bk.scatter([fakeScatter[i]], [fakeScatter[i]], size = [legendCircles[i]], legend = int(legendCircles[i]), color = 'red', fill_line = None, alpha = 0.4)
# Disable background grid
bk.grid().grid_line_color = None
#print centers
### Legend
#x = -74.25
#y = 40.8
#height = 0.003
#legend_box_y = y + 0.5 * height * len(legendColors)
#bk.rect([x+0.032], [legend_box_y], width = 0.12, height = height*len(legendColors)*1.15, color = "#ffffff", line_color = "gray")
##x = -74.25
##y = 40.9
#for color in legendColors:
## #print "Color: ", a
## #print "x:", x
## #print "y:", y
##
# bk.rect([x], [y], color = color, width=0.03, height=height)
# #bk.text([x], [y], text = agency, angle=0, text_font_size="7pt", text_align="center", text_baseline="middle")
# y = y + height
#legend_bottom_y = 40.797
#legend_top_y = legend_bottom_y + 0.92 * height * len(legendColors)
#bk.text([-74.225], [legend_bottom_y], text = agency2, angle = 0, text_font_size = "7pt", text_align = "left")
#bk.text([-74.225], [legend_top_y], text = agency1, angle = 0, text_font_size = "7pt", text_align = "left")
#bokeh.embed.components(fig, bokeh.resources.CDN)
bk.show()
tgt_explorations.append((exploration, feedback))
tgt_dataset = {
'm_channels': ex.m_channels,
's_channels': ex.s_channels,
'explorations': tgt_explorations
}
# Graphs
try:
from bokeh import plotting
plotting.output_file('goal_reuse.html')
xs = [explo[1]['s_signal']['x'] for explo in dataset['explorations']]
ys = [explo[1]['s_signal']['y'] for explo in dataset['explorations']]
plotting.scatter(xs,
ys,
x_range=[-1, 1],
y_range=[-1, 1],
title='explorers library: goal reuse algorithm example',
fill_alpha=0.5,
line_color=None,
radius=2.0,
radius_units='screen')
plotting.show()
except ImportError:
print(
'exploration went fine, but you need the bokeh library (http://bokeh.pydata.org/) to display it'
)
# Graph
# the red dots are from motor babbling
# the blue dots from goal babbling
try:
from bokeh import plotting
plotting.output_file('goal_babbling.html')
xs = [explo[1]['s_signal']['x'] for explo in dataset['explorations']]
ys = [explo[1]['s_signal']['y'] for explo in dataset['explorations']]
plotting.scatter(xs,
ys,
x_range=[-1, 1],
y_range=[-1, 1],
title='explorers library: exploration example',
fill_alpha=0.5,
line_color=None,
radius=2.0,
radius_units='screen')
plotting.hold(True)
plotting.scatter(xs[:ex_cfg.eras[0]],
ys[:ex_cfg.eras[0]],
x_range=[-1, 1],
y_range=[-1, 1],
title='explorers library: exploration example',
fill_alpha=0.5,
color='red',
line_color=None,
radius=2.0,
radius_units='screen')
def drawPlot(shapeFilename, zipBorough, zipCount):
# Read the ShapeFile
dat = shapefile.Reader(shapeFilename)
# Creates a dictionary for zip: {lat_list: [], lng_list: []}.
zipCodes = []
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 zipBorough: # was in zipBorough:
# Gets shape for this zip.
shape = dat.shapeRecord(record_index).shape
### Only first polygon if non-contiguous zip codes
#firstPoint = shape.parts[0]
#if len(shape.parts) > 1:
# lastPoint = shape.parts[1]
#else:
# lastPoint = len(shape.points)
#zipCodes.append(currentZip)
#points = shape.points[firstPoint : lastPoint]
#
## 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)
### All shapes for each zip code
for part in range(len(shape.parts)):
zipCodes.append(currentZip)
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
# Creates the Plot
bk.output_file("problem4.html")
bk.hold()
north = 40.915
east = -73.651
south = 40.496
west = -74.256
width = east - west
height = north - south
x_range = [west - 0.05 * width, east + 0.05 * width]
y_range = [south - 0.05 * height, north + 0.05 * height]
TOOLS = "pan,wheel_zoom,box_zoom,reset,previewsave"
fig = bk.figure(title="311 Complaints by Zip Code", \
tools=TOOLS, background_fill = "#f8f8f8", x_range = x_range, y_range = y_range, plot_height=800, plot_width = 1200)
#circleSizes = [zipCount[zipCode] for zipCode in zipCodes]
circleSizes = []
for zipCode in zipCodes:
if zipCode in zipCount:
circleSizes.append(zipCount[zipCode])
else:
circleSizes.append(1) # for taking logs
#print circleSizes
#print "aaaaaaaaaaaaaaaaaaaaa: ", len(polygons['lng_list']), len(circleSizes)
logCircleSizes = ((np.log(np.array(circleSizes)))**3.7) * 0.009
#print logCircleSizes
logCircleSizes = list(logCircleSizes)
#circleSizes = [x for x in range(len(zipCodes))]
# Creates the polygons
bk.patches(polygons['lng_list'],
polygons['lat_list'],
fill_color="#fee8c8",
line_color="gray")
bk.scatter(polygons['centerLon_list'],
polygons['centerLat_list'],
size=logCircleSizes,
alpha=0.4,
line_color=None,
fill_color='red')
#circleSizes = [0.009 * (size ** 3.7) for size in circleSizes]
#bk.scatter(centers['lon'], centers['lat'], size = circleSizes, alpha = 0.4, line_color = None, fill_color = 'red')
#circleSizeArr = np.array(circleSizes)
#maxSize = np.max(circleSizeArr)
#circleSizeArr[circleSizeArr == 0] = 100
#minSize = np.min(circleSizeArr)
#legendN = 5
#legendCircles = list(np.linspace(minSize, maxSize, legendN))
#legendCounts = [str(int((size ** (1.0/3.7))/0.009)) + " complaints" for size in legendCircles]
#fakeScatter = [0 for x in range(legendN)]
#fakeScatterX = [-74.23 for x in range(legendN)]
#fakeScatterXtext = [-74.23 + 0.05 for x in range(legendN)]
#fakeScatterY = [40.8, 40.8 + 0.02, 40.8 + 0.036, 40.8 + 0.056,40.8 + 0.083]
#bk.scatter(fakeScatterX, fakeScatterY, size = legendCircles, fill_color = 'red', line_color = None, alpha = 0.4)
#bk.text(fakeScatterXtext[1:], fakeScatterY[1:], text = legendCounts[1:], angle = 0, text_align = "left", text_font_size = "7pt", text_baseline = 'middle')
# Disable background grid
bk.grid().grid_line_color = None
bk.show()
