def _create_plot_component():
red = util.Color(255, 0, 0, 255)
green = util.Color(0, 255, 0, 255)
blue = util.Color(0, 0, 255, 255)
purple = util.Color(125, 0, 255, 255)
white = util.Color(255, 255, 255, 255)
black = util.Color(0, 0, 0, 255)
clear = util.Color(0, 0, 0, 0)
with Timer("Loeading") as arTimer:
#glyphs = npg.load_csv("../data/circlepoints.csv", 1, 2, 3, 4)
#glyphs = npg.load_hdf("../data/CensusTracts.hdf5", "__data__", "LAT", "LON")
glyphs = npg.load_hdf("../data/tweets-subset.hdf",
"test",
"longitude",
"latitude",
vc="lang_primary")
screen = (800, 600)
ivt = util.zoom_fit(screen, glyphs.bounds())
with Timer("Abstract-Render") as arTimer:
image = core.render(
glyphs, infos.encode(["Arabic", "English", "Turkish", "Russian"]),
npg.PointCountCategories(),
npg.Spread(2) + categories.HDAlpha(
[red, blue, green, purple, black], alphamin=.3, log=True),
screen, ivt)
# image = core.render(glyphs,
# infos.valAt(4,0),
# npg.PointCount(),
# npg.Spread(1) + numeric.BinarySegment(white, black, 1),
# screen,
# ivt)
# Create a plot data object and give it this data
pd = ArrayPlotData()
pd.set_data("imagedata", image)
# Create the plot
plot = Plot(pd)
img_plot = plot.img_plot("imagedata")[0]
# Tweak some of the plot properties
plot.title = "Abstract Rendering"
plot.padding = 50
return plot
def _create_plot_component():
red = util.Color(255,0,0,255)
green = util.Color(0,255,0,255)
blue = util.Color(0,0,255,255)
purple = util.Color(125,0,255,255)
white = util.Color(255,255,255,255)
black = util.Color(0,0,0,255)
clear = util.Color(0,0,0,0)
with Timer("Loeading") as arTimer:
#glyphs = npg.load_csv("../data/circlepoints.csv", 1, 2, 3, 4)
glyphs = npg.load_hdf("../data/CensusTracts.hdf5", "__data__", "LAT", "LON")
#glyphs = npg.load_hdf("../data/tweets-subset.hdf", "test",
# "longitude", "latitude", vc="lang_primary")
screen = (800,600)
ivt = util.zoom_fit(screen,glyphs.bounds())
with Timer("Abstract-Render") as arTimer:
image = core.render(glyphs,
infos.encode(["Arabic","English","Turkish","Russian"]),
npg.PointCountCategories(),
npg.Spread(2) + categories.HDAlpha([red, blue, green, purple, black], alphamin=.3, log=True),
screen,
ivt)
# image = core.render(glyphs,
# infos.valAt(4,0),
# npg.PointCount(),
# npg.Spread(1) + numeric.BinarySegment(white, black, 1),
# screen,
# ivt)
# Create a plot data object and give it this data
pd = ArrayPlotData()
pd.set_data("imagedata", image)
# Create the plot
plot = Plot(pd)
img_plot = plot.img_plot("imagedata")[0]
# Tweak some of the plot properties
plot.title = "Abstract Rendering"
plot.padding = 50
return plot
def _create_plot_component():
red = util.Color(255,0,0,255)
green = util.Color(0,255,0,255)
blue = util.Color(0,0,255,255)
white = util.Color(255,255,255,255)
black = util.Color(0,0,0,255)
shape = glyphset.ShapeCodes.POINT
#glyphs = glyphset.load_csv("../data/checkerboard.csv", 2, 0, 1, 3,1,1, shape)
#glyphs = glyphset.load_csv("../data/circlepoints.csv", 1, 2, 3, 4,.1,.1, shape)
#glyphs = glyphset.load_csv("../data/sourceforge.csv", 1, 1, 2, -1,.1,.1, shape)
glyphs = glyphset.load_hdf("../data/CensusTracts.hdf5", "__data__", "LON", "LAT", None, .1, .1, shape)
#glyphs = glyphset.load_hdf("../data/tweets-subset.hdf", "test", "longitude", "latitude", None, .1, .1, shape)
screen = (800,600)
ivt = util.zoom_fit(screen,glyphs.bounds())
with Timer("Abstract-Render") as arTimer:
# image = core.render(glyphs,
# infos.val(),
# categories.CountCategories(),
# categories.HDAlpha([red, blue]),
# screen,
# ivt)
image = core.render(glyphs,
infos.valAt(4,0),
numeric.Count(),
numeric.BinarySegment(white, black, 1),
screen,
ivt)
print("screen x image -- {0} x {1}".format(screen, image.shape))
# Create a plot data object and give it this data
pd = ArrayPlotData()
pd.set_data("imagedata", image)
# Create the plot
plot = Plot(pd)
img_plot = plot.img_plot("imagedata")[0]
# Tweak some of the plot properties
plot.title = "Abstract Rendering"
plot.padding = 50
return plot
def _create_plot_component():
red = util.Color(255, 0, 0, 255)
green = util.Color(0, 255, 0, 255)
blue = util.Color(0, 0, 255, 255)
purple = util.Color(125, 0, 255, 255)
white = util.Color(255, 255, 255, 255)
black = util.Color(0, 0, 0, 255)
shape = glyphset.ShapeCodes.RECT
glyphs = blzg.load_csv(
"../data/circlepoints.csv",
"x",
"y",
"series",
schema="{r:float32, theta:float32, x:float32, y:float32, series:int32}"
)
screen = (800, 600)
ivt = util.zoom_fit(screen, glyphs.bounds())
with Timer("Abstract-Render") as arTimer:
image = core.render(
glyphs, infos.val(), blzg.CountCategories("int32"),
categories.HDAlpha([red, blue, green, purple, black]), screen, ivt)
# image = core.render(glyphs,
# infos.valAt(4,0),
# blzg.Count(),
# numeric.BinarySegment(white, black, 1),
# screen,
# ivt)
# Create a plot data object and give it this data
pd = ArrayPlotData()
pd.set_data("imagedata", image)
# Create the plot
plot = Plot(pd)
img_plot = plot.img_plot("imagedata")[0]
# Tweak some of the plot properties
plot.title = "Abstract Rendering"
plot.padding = 50
return plot
def _create_plot_component():
red = util.Color(255,0,0,255)
green = util.Color(0,255,0,255)
blue = util.Color(0,0,255,255)
purple = util.Color(125,0,255,255)
white = util.Color(255,255,255,255)
black = util.Color(0,0,0,255)
shape = glyphset.ShapeCodes.RECT
glyphs = blzg.load_csv("../data/circlepoints.csv", "x", "y", "series",
schema="{r:float32, theta:float32, x:float32, y:float32, series:int32}")
screen = (800,600)
ivt = util.zoom_fit(screen,glyphs.bounds())
with Timer("Abstract-Render") as arTimer:
image = core.render(glyphs,
infos.val(),
blzg.CountCategories("int32"),
categories.HDAlpha([red, blue, green, purple, black]),
screen,
ivt)
# image = core.render(glyphs,
# infos.valAt(4,0),
# blzg.Count(),
# numeric.BinarySegment(white, black, 1),
# screen,
# ivt)
# Create a plot data object and give it this data
pd = ArrayPlotData()
pd.set_data("imagedata", image)
# Create the plot
plot = Plot(pd)
img_plot = plot.img_plot("imagedata")[0]
# Tweak some of the plot properties
plot.title = "Abstract Rendering"
plot.padding = 50
return plot
def downsample(data, transform, plot_state):
screen_size = [span(plot_state['screen_x']),
span(plot_state['screen_y'])]
scale_x = span(plot_state['data_x'])/float(span(plot_state['screen_x']))
scale_y = span(plot_state['data_y'])/float(span(plot_state['screen_y']))
#How big would a full plot of the data be at the current resolution?
plot_size = [screen_size[0] / scale_x, screen_size[1] / scale_y]
glyphspec = transform['glyphspec']
xcol = glyphspec['x']['field']
ycol = glyphspec['y']['field']
size = glyphspec['size']['default'] ##TODO: Will not work for data-derived sizes...
###Translate the resample paramteres to server-side rendering....
###TODO: Should probably handle this type-based-unpacking server_backend so downsamples get a consistent view of the data
if type(data) is dict:
xcol = data[xcol]
ycol = data[ycol]
else:
table = data.select(columns=[xcol, ycol])
xcol = table[xcol]
ycol = table[ycol]
shaper = _shaper(glyphspec['type'], size)
glyphs = glyphset.Glyphset([xcol, ycol], ar.EmptyList(), shaper, colMajor=True)
bounds = glyphs.bounds()
ivt = ar.zoom_fit(plot_size, bounds, balanced=False)
image = ar.render(glyphs,
transform['info'].reify(),
transform['agg'].reify(),
transform['shader'].reify(),
plot_size, ivt)
return {'image': [image],
'x': [0],
'y': [0],
'dw': [image.shape[0]],
'dh': [image.shape[1]],
}
