def gen_real_view(self, resetting, x_range, y_range):
if x_range is None:
x_range = self.xrange
if y_range is None:
y_range = self.yrange0
# ipdb.set_trace()
if resetting:
return dynspread(datashade(self.real_curve,
cmap='blue',
normalization='linear',
x_range=self.xrange,
y_range=self.yrange0,
min_alpha=self.min_alpha,
dynamic=False),
max_px=max_px)
else:
return dynspread(datashade(self.real_curve,
cmap='blue',
normalization='linear',
x_range=x_range,
y_range=y_range,
min_alpha=self.min_alpha,
dynamic=False),
max_px=max_px)
def plot_phase_diagrams(filenames,fileout):
for i in range(len(filenames)):
print 'i: ',i
ds = yt.load(filenames[i])
center_guess = initial_center_guess(ds,track_name)
halo_center = get_halo_center(ds,center_guess)
rb = sym_refine_box(ds,halo_center)
args = filenames[i].split('/')
sim_label = args[-3]
dens = np.log10(rb['H_nuclei_density'])
temp = np.log10(rb['Temperature'])
df = pd.DataFrame({'temp':temp, 'dens':dens})
phase_scatter = hv.Scatter(df,kdims=['dens'],vdims=['temp'],label=sim_label)
#phase_data = np.zeros((len(rb['H_nuclei_density']),2))
#phase_data[:,0] = np.log10(rb['H_nuclei_density'])
#phase_data[:,1] = np.log10(rb['Temperature'])
#points = hv.Points(phase_data,kdims=['nH','temp'],label=sim_label)
hv.opts({'Histogram': {'style': {'alpha':0.3, 'fill_color':'k'}}})
xhist = (histogram(phase_scatter, bin_range=(-7.5, 1), dimension='dens',normed=True)) #,alpha=0.3, fill_color='k'))
yhist = (histogram(phase_scatter, bin_range=(3, 8.5), dimension='temp',normed=True)) #,alpha=0.3, fill_color='k'))
if i == 0:
phase_plot = (datashade(phase_scatter,cmap=cm.plasma, dynamic=False,x_range=(-7.5,1),y_range=(3,8.5))) << yhist(plot=dict(width=125)) << xhist(plot=dict(height=125))
else:
plot2 = (datashade(phase_scatter,cmap=cm.plasma, dynamic=False,x_range=(-7.5,1),y_range=(3,8.5))) << yhist(plot=dict(width=125)) << xhist(plot=dict(height=125))
phase_plot = phase_plot + plot2
renderer = hv.renderer('bokeh').instance(fig='html')
renderer.save(phase_plot, fileout)
return
def grid(self, tiles=False, **kwargs):
x = kwargs.get("x", self._obj.SCHISM_hgrid_node_x[:].values)
y = kwargs.get("y", self._obj.SCHISM_hgrid_node_y[:].values)
tes = kwargs.get("tri3", self._obj.SCHISM_hgrid_face_nodes.values)
width = kwargs.get("width", 800)
height = kwargs.get("height", 600)
opts.defaults(opts.WMTS(width=width, height=height))
tile = gv.WMTS("https://b.tile.openstreetmap.org/{Z}/{X}/{Y}.png")
nodes = pd.DataFrame({"longitude": x, "latitude": y})
points = gv.operation.project_points(gv.Points(nodes))
if tes.shape[1] == 3:
elems = pd.DataFrame(tes, columns=["a", "b", "c"])
trimesh = gv.TriMesh((elems, points)).edgepaths
if tiles:
return tile * datashade(
trimesh, precompute=True, cmap=["black"])
else:
return datashade(trimesh, precompute=True,
cmap=["green"]).opts(width=width,
height=height)
else: # there are quads
elems = pd.DataFrame(tes, columns=["a", "b", "c", "d"])
quads = elems.loc[~elems.d.isna()].copy()
quads = quads.reset_index(drop=True)
ap = nodes.loc[quads.a, ["longitude", "latitude"]]
bp = nodes.loc[quads.b, ["longitude", "latitude"]]
cp = nodes.loc[quads.c, ["longitude", "latitude"]]
dp = nodes.loc[quads.d, ["longitude", "latitude"]]
quads["ap"] = ap.values.tolist()
quads["bp"] = bp.values.tolist()
quads["cp"] = cp.values.tolist()
quads["dp"] = dp.values.tolist()
q = gv.Polygons([
quads.loc[i, ["ap", "bp", "cp", "dp"]].tolist()
for i in range(quads.shape[0])
]).options(fill_alpha=0, line_color="black")
triangles = elems.loc[elems.d.isna()]
trimesh = gv.TriMesh((triangles, points)).edgepaths
g1 = datashade(trimesh, precompute=True, cmap=["black"])
if tiles:
return tile * g1 * q
else:
return g1 * q
def view_datashade(self):
"""
Use of datashade for performance and line for hover tool capabilities
:return: Panel of this combination
"""
# Select only sufficient data
if self.x in self.y:
self.y.remove(self.x)
if self.y == []:
return self.gif
df = self.dataframe[[self.x] + self.y].copy()
plot_opts = {
'Scatter': {
'color': self.color_key,
'marker': self.marker_keys,
'size': 10
},
'Curve': {
'color': self.color_key
}
}
lines_overlay = df.hvplot.scatter(
**self.plot_options).options(plot_opts)
def hover_curve(x_range=[df.index.min(),
df.index.max()]): # , y_range):
# Compute
dataframe = df.copy()
if x_range is not None:
dataframe = dataframe[(dataframe[self.x] > x_range[0])
& (dataframe[self.x] < x_range[1])]
data_length = len(dataframe) * len(dataframe.columns)
step = 1 if data_length < self.max_step else data_length // self.max_step
plot_df = dataframe[::step].hvplot.line(**self.plot_options) * \
dataframe[::step*60].hvplot.scatter(**self.plot_options)
plot_opts = {
'Scatter': {
'color': 'k',
'marker': self.marker_keys,
'size': 10
},
'Curve': {
'color': self.color_key
}
}
if len(self.y) != 1:
plot_opts['Scatter']['color'] = self.color_key
return plot_df.options(plot_opts)
# Define a RangeXY stream linked to the image
rangex = hv.streams.RangeX(source=lines_overlay)
data_shade_plot = hv.DynamicMap(hover_curve, streams=[rangex])
if len(self.y) == 1:
data_shade_plot *= datashade(lines_overlay)
else:
data_shade_plot *= datashade(lines_overlay,
aggregator=ds.count_cat('Variable'))
return pn.panel(data_shade_plot)
def _get_points(self):
embeddings = self.embeddings
classes = self.classes
if (self.label_flag) and (classes is not None):
data = pd.DataFrame(embeddings)
data.columns = ['ivis 1', 'ivis 2']
data['label'] = classes
num_ks = len(np.unique(classes))
color_key = list(enumerate(Sets1to3[0:num_ks]))
embed = {
k: hv.Points(data.values[classes == k, :],
['ivis 1', 'ivis 2'],
'k',
label=str(k)).opts(color=v, size=0)
for k, v in color_key
}
dse = dynspread(
datashade(hv.NdOverlay(embed, kdims=['k']),
aggregator=ds.by('k', ds.count())))
color_points = hv.NdOverlay({
k: hv.Points([0, 0]).opts(color=v, size=0)
for k, v in color_key
})
points = color_points * dse
else:
points = datashade(hv.Points(embeddings))
points.opts(height=400, width=500, xaxis=None, yaxis=None)
return points
def test_datashade_in_overlay_selection(self):
points = Points(self.data)
layout = points * dynspread(datashade(points))
lnk_sel = link_selections.instance(unselected_color='#ff0000')
linked = lnk_sel(layout)
current_obj = linked[()]
# Check base points layer
self.check_base_points_like(current_obj[()].Points.I, lnk_sel)
# Check selection layer
self.check_overlay_points_like(current_obj[()].Points.II, lnk_sel,
self.data)
# Check RGB base layer
self.assertEqual(
current_obj[()].RGB.I,
dynspread(
datashade(points, cmap=lnk_sel.unselected_cmap,
alpha=255))[()])
# Check RGB selection layer
self.assertEqual(
current_obj[()].RGB.II,
dynspread(datashade(points, cmap=lnk_sel.selected_cmap,
alpha=255))[()])
# Perform selection of second and third point
selectionxy = TestLinkSelections.get_value_with_key_type(
lnk_sel._selection_expr_streams,
hv.Points).input_streams[0].input_stream.input_streams[0]
self.assertIsInstance(selectionxy, SelectionXY)
selectionxy.event(bounds=(0, 1, 5, 5))
current_obj = linked[()]
# Check that base points layer is unchanged
self.check_base_points_like(current_obj[()].Points.I, lnk_sel)
# Check points selection layer
self.check_overlay_points_like(current_obj[()].Points.II, lnk_sel,
self.data.iloc[1:])
# Check that base RGB layer is unchanged
self.assertEqual(
current_obj[()].RGB.I,
dynspread(
datashade(points, cmap=lnk_sel.unselected_cmap,
alpha=255))[()])
# Check selection RGB layer
self.assertEqual(
current_obj[()].RGB.II,
dynspread(
datashade(points.iloc[1:],
cmap=lnk_sel.selected_cmap,
alpha=255))[()])
def grid(self, tiles=False, **kwargs):
x = kwargs.get('x',self._obj.SCHISM_hgrid_node_x[:].values)
y = kwargs.get('y',self._obj.SCHISM_hgrid_node_y[:].values)
tes = kwargs.get('tri3',self._obj.SCHISM_hgrid_face_nodes.values)
width = kwargs.get('width', 800)
height = kwargs.get('height', 600)
opts.defaults(opts.WMTS(width=width, height=height))
tile = gv.WMTS('https://b.tile.openstreetmap.org/{Z}/{X}/{Y}.png')
nodes = pd.DataFrame({'longitude':x,'latitude':y})
points = gv.operation.project_points(gv.Points(nodes))
if tes.shape[1] == 3 :
elems = pd.DataFrame(tes, columns=['a', 'b', 'c'])
trimesh=gv.TriMesh((elems, points)).edgepaths
if tiles:
return tile * datashade(trimesh, precompute=True, cmap=['black'])
else:
return datashade(trimesh, precompute=True, cmap=['green']).opts(width=width,height=height)
else: # there are quads
elems = pd.DataFrame(tes, columns=['a', 'b', 'c', 'd'])
quads = elems.loc[~elems.d.isna()].copy()
quads = quads.reset_index(drop=True)
ap = nodes.loc[quads.a,['longitude','latitude']]
bp = nodes.loc[quads.b,['longitude','latitude']]
cp = nodes.loc[quads.c,['longitude','latitude']]
dp = nodes.loc[quads.d,['longitude','latitude']]
quads['ap'] = ap.values.tolist()
quads['bp'] = bp.values.tolist()
quads['cp'] = cp.values.tolist()
quads['dp'] = dp.values.tolist()
q = gv.Polygons([quads.loc[i,['ap','bp','cp','dp']].tolist() for i in range(quads.shape[0])]).options(fill_alpha=0, line_color='black')
triangles = elems.loc[elems.d.isna()]
trimesh=gv.TriMesh((triangles, points)).edgepaths
g1 = datashade(trimesh, precompute=True, cmap=['black'])
if tiles :
return tile * g1 * q
else:
return g1 * q
def test_datashade_selection(self):
scatter = hv.Scatter(self.data, kdims='x', vdims='y')
layout = scatter + dynspread(datashade(scatter))
lnk_sel = link_selections.instance()
linked = lnk_sel(layout)
current_obj = linked[()]
# Check base scatter layer
self.check_base_scatter_like(current_obj[0][()].Scatter.I, lnk_sel)
# Check selection layer
self.check_overlay_scatter_like(current_obj[0][()].Scatter.II, lnk_sel,
self.data)
# Check RGB base layer
self.assertEqual(
current_obj[1][()].RGB.I,
dynspread(
datashade(scatter, cmap=lnk_sel.unselected_cmap,
alpha=255))[()])
# Check RGB selection layer
self.assertEqual(
current_obj[1][()].RGB.II,
dynspread(datashade(scatter, cmap=lnk_sel.selected_cmap,
alpha=0))[()])
# Perform selection of second and third point
boundsxy = lnk_sel._selection_expr_streams[0]._source_streams[0]
self.assertIsInstance(boundsxy, hv.streams.BoundsXY)
boundsxy.event(bounds=(0, 1, 5, 5))
current_obj = linked[()]
# Check that base scatter layer is unchanged
self.check_base_scatter_like(current_obj[0][()].Scatter.I, lnk_sel)
# Check scatter selection layer
self.check_overlay_scatter_like(current_obj[0][()].Scatter.II, lnk_sel,
self.data.iloc[1:])
# Check that base RGB layer is unchanged
self.assertEqual(
current_obj[1][()].RGB.I,
dynspread(
datashade(scatter, cmap=lnk_sel.unselected_cmap,
alpha=255))[()])
# Check selection RGB layer
self.assertEqual(
current_obj[1][()].RGB.II,
dynspread(
datashade(scatter.iloc[1:],
cmap=lnk_sel.selected_cmap,
alpha=255))[()])
def holoviews_radial_profiles(weight_by=None):
dens = np.log10(rb['H_nuclei_density'])
temp = np.log10(rb['Temperature'])
Zgas = np.log10(rb['metallicity'])
cell_mass = rb['cell_mass'].in_units('Msun')
cell_volume = rb['cell_volume'].in_units('kpc**3')
x = rb['x']
y = rb['y']
z = rb['z']
halo_center = ds.arr(rb_center,'code_length')
dist = np.sqrt((halo_center[0]-rb['x'])**2.+(halo_center[1]-rb['y'])**2.+(halo_center[2]-rb['z'])**2.).in_units('kpc')
df = pd.DataFrame({'temp':temp, 'dens':dens, 'Zgas':Zgas,'cell_volume':cell_volume,
'x':x,'y':y,'z':z,'dist':dist,'cell_mass':cell_mass})
temp_dist = hv.Scatter(df,kdims=['dist'],vdims=['temp'],label="Temperature ")
dens_dist = hv.Scatter(df,kdims=['dist'],vdims=['dens'],label='Hydrogen Number Density')
metal_dist = hv.Scatter(df,kdims=['dist'],vdims=['Zgas'],label='Metallicity')
if weight_by == None:
dist_plots = (datashade(temp_dist,cmap=cm.Reds, dynamic=False,x_range=(0,60),y_range=(2,8.4)).opts(plot=dict(aspect='square'))
+ datashade(dens_dist,cmap=cm.Blues, dynamic=False,x_range=(0,60),y_range=(-6.5,2)).opts(plot=dict(aspect='square'))
+ datashade(metal_dist,cmap=cm.BuGn, dynamic=False,x_range=(0,60),y_range=(-8.5,1.4)).opts(plot=dict(aspect='square')))
fileout= 'basic_profile_'+args[-3]+'_'+args[-1]
if weight_by == 'cell_mass':
temp_shade = aggregate(hv.Scatter(df,['dist','temp']),y_range=(2,8.4),aggregator=dshade.sum('cell_mass'))
temp_shade = temp_shade.opts(plot=dict(colorbar=True,aspect='square',logz=True),style=dict(cmap=cm.Reds))
dens_shade = aggregate(hv.Scatter(df,['dist','dens']),y_range=(-7,2.5),aggregator=dshade.sum('cell_mass'))
dens_shade = dens_shade.opts(plot=dict(colorbar=True,aspect='square',logz=True),style=dict(cmap=cm.Blues))
metal_shade = aggregate(hv.Scatter(df,['dist','Zgas']),y_range=(-7,2.5),aggregator=dshade.sum('cell_mass'))
metal_shade = metal_shade.opts(plot=dict(colorbar=True,aspect='square',logz=True),style=dict(cmap=cm.BuGn))
dist_plots = (temp_shade + dens_shade + metal_shade)
fileout = 'basic_profile_cell_mass_'+args[-3]+'_'+args[-1]
if weight_by == 'cell_volume':
temp_shade = aggregate(hv.Scatter(df,['dist','temp']),y_range=(2,8.4),aggregator=dshade.sum('cell_volume'))
temp_shade = temp_shade.opts(plot=dict(colorbar=True,aspect='square',logz=True),style=dict(cmap=cm.Reds))
dens_shade = aggregate(hv.Scatter(df,['dist','dens']),y_range=(-7,2.5),aggregator=dshade.sum('cell_volume'))
dens_shade = dens_shade.opts(plot=dict(colorbar=True,aspect='square',logz=True),style=dict(cmap=cm.Blues))
metal_shade = aggregate(hv.Scatter(df,['dist','Zgas']),y_range=(-7,2.5),aggregator=dshade.sum('cell_volume'))
metal_shade = metal_shade.opts(plot=dict(colorbar=True,aspect='square',logz=True),style=dict(cmap=cm.BuGn))
dist_plots = (temp_shade + dens_shade + metal_shade)
fileout = 'basic_profile_cell_vol_'+args[-3]+'_'+args[-1]
renderer = Store.renderers['matplotlib'].instance(fig='pdf', holomap='gif')
renderer.save(dist_plots, fileout)
return
def plot_performance(ds_preds, full=False):
"""Multiple plots using xr_preds"""
p = hv_plot_prediction(ds_preds.isel(t_source=10))
display(p)
n = len(ds_preds.t_source)
d_ahead = ds_preds.mean(['t_source'])['nll'].groupby('t_ahead_hours').mean()
nll_vs_tahead = (hv.Curve(
(d_ahead.t_ahead_hours,
d_ahead)).redim(x='hours ahead',
y='nll').opts(
title=f'NLL vs time ahead (no. samples={n})'))
display(nll_vs_tahead)
# Make a plot of the NLL over time. Does this solution get worse with time?
if full:
d_source = ds_preds.mean(['t_ahead'])['nll'].groupby('t_source').mean()
nll_vs_time = (hv.Curve(d_source).opts(
title='Error vs time of prediction'))
display(nll_vs_time)
# A scatter plot is easy with xarray
if full:
tlim = (ds_preds.y_true.min().item(), ds_preds.y_true.max().item())
true_vs_pred = datashade(hv.Scatter(
(ds_preds.y_true,
ds_preds.y_pred))).redim(x='true', y='pred').opts(width=400,
height=400,
xlim=tlim,
ylim=tlim,
title='Scatter plot')
true_vs_pred = dynspread(true_vs_pred)
true_vs_pred
display(true_vs_pred)
def make_view(self, x_range=None, y_range=None, **kwargs):
#map_tiles = tiles.opts(style=dict(alpha=self.alpha), plot=options)
points = hv.Points(
df,
kdims=['X_CORD', 'Y_CORD'],
vdims=['EVENT_COUNT', 'EventType', 'SUB', 'day', 'FEEDER_ID'])
if (self.BySUB & self.ByFeeder & self.ByDay):
selected = points.select(EventType=self.plot,
EVENT_COUNT=self.numEvents,
SUB=self.Substations,
day=self.DayNum,
FEEDER_ID=self.Feeder)
elif (self.BySUB & self.ByFeeder & (not self.ByDay)):
selected = points.select(EventType=self.plot,
EVENT_COUNT=self.numEvents,
SUB=self.Substations,
FEEDER_ID=self.Feeder)
elif (self.BySUB & (not self.ByFeeder) & self.ByDay):
selected = points.select(EventType=self.plot,
EVENT_COUNT=self.numEvents,
SUB=self.Substations,
day=self.DayNum)
elif (self.BySUB & (not self.ByFeeder) & (not self.ByDay)):
selected = points.select(EventType=self.plot,
EVENT_COUNT=self.numEvents,
SUB=self.Substations)
elif ((not self.BySUB) & self.ByFeeder & self.ByDay):
selected = points.select(EventType=self.plot,
EVENT_COUNT=self.numEvents,
day=self.DayNum,
FEEDER_ID=self.Feeder)
elif ((not self.BySUB) & self.ByFeeder & (not self.ByDay)):
selected = points.select(EventType=self.plot,
EVENT_COUNT=self.numEvents,
FEEDER_ID=self.Feeder)
elif ((not self.BySUB) & (not self.ByFeeder) & self.ByDay):
selected = points.select(EventType=self.plot,
EVENT_COUNT=self.numEvents,
day=self.DayNum)
else:
selected = points.select(EventType=self.plot,
EVENT_COUNT=self.numEvents)
SagSwellPts = datashade(selected,
x_sampling=1,
y_sampling=1,
cmap=self.colormap,
dynamic=False,
x_range=x_range,
y_range=y_range,
width=640,
height=380)
dsss = dynspread(SagSwellPts,
shape='circle',
max_px=self.maxpix,
threshold=self.threshhold)
#return map_tiles * dsss
return dsss
def view(self):
if datashaded and self.nodeCount > 1:
plot = dynspread(
datashade(self.plot,
normalization='linear',
width=self.size,
height=self.size,
cmap=self.colorMap[self.color_palette]))
self.points.opts(cmap=self.colorMap[self.color_palette],
color=self.node_color,
size=self.node_size)
return (plot, self.points)
else:
if (min(self.plot.data['weight']) < max(
self.plot.data['weight'])):
self.points.opts(cmap=self.colorMap[self.color_palette],
color=self.node_color,
size=self.node_size)
self.plot.opts(
edge_cmap=self.colorMap[self.color_palette],
edge_color='weight',
edge_line_width=dim('weight').norm() * 5 + 0.1,
edge_line_alpha=0.1 + dim('weight').norm() * 0.9)
else:
self.points.opts(cmap=self.colorMap[self.color_palette],
color=self.node_color,
size=self.node_size)
self.plot.opts(edge_cmap=self.colorMap[self.color_palette],
edge_color=dim('weight'),
edge_line_width=dim('weight') * 3,
edge_line_alpha=dim('weight'))
return (self.plot, self.points)
def _process(self, element, key=None):
vdim = self.p.vdim
if self.p.aggregator == 'mean':
aggregator = ds.mean(vdim)
elif self.p.aggregator == 'std':
aggregator = ds.std(vdim)
elif self.p.aggregator == 'count':
aggregator = ds.count()
kwargs = dict(cmap=cc.palette[self.p.cmap], aggregator=aggregator)
if self.p.width is not None:
kwargs.update(width=self.p.width,
height=self.p.height,
streams=[hv.streams.RangeXY])
datashaded = dynspread(datashade(element, **kwargs))
# decimate_opts = dict(plot={'tools':['hover', 'box_select']},
# style={'alpha':0, 'size':self.p.decimate_size,
# 'nonselection_alpha':0})
# decimated = decimate(element, max_samples=self.p.max_samples).opts(**decimate_opts)
return datashaded # * decimated
def scatter_dist_by_mappings(dataset, x_kdims, y_kdims,
mappings,
selection_dim="Gene",
datashade_=False,
dynspread_=False,
):
data_groups = {name: dataset.sel({selection_dim: genes}) for name, genes in mappings.items()}
data_group_dfs = {k: v[[x_kdims, y_kdims]].to_dataframe() for k, v in data_groups.items()}
points = {k: hv.Points(val, kdims=[x_kdims, y_kdims]) for k, val in data_group_dfs.items()}
dist_x = {k: univariate_kde(hv.Distribution(p, kdims=[y_kdims], group="dist_x"), n_samples=1000)
for k, p in points.items()}
dist_y = {k: univariate_kde(hv.Distribution(p, kdims=[x_kdims], group="dist_y"), n_samples=1000)
for k, p in points.items()}
if datashade_:
points_overlay = datashade(hv.NdOverlay(points))
if dynspread_:
points_overlay = dynspread(points_overlay)
else:
points_overlay = hv.NdOverlay(points)
return points_overlay << hv.NdOverlay(dist_x) << hv.NdOverlay(dist_y)
def shade(hv_obj, reduction='any', color=None, spread=False):
"""
Apply datashading to a holoviews object.
hv_obj: a holovies object like Curve, Scatter, etc.
reduction: Most common will be 'any' and 'count'.
Supply any name here to see list of valid reductions
color: Mostly used for 'any' aggregation to specify a color
spread: Smear out points slightly bigger than 1 pixel for easier
visibility
"""
import datashader as ds
from holoviews.operation.datashader import datashade, dynspread
if reduction not in ALLOWED_REDUCTIONS:
raise ValueError(
'Allowed reductions are {}'.format(ALLOWED_REDUCTIONS))
reducer = getattr(ds.reductions, reduction)
kwargs = dict(aggregator=reducer())
if color is None and reduction == 'any':
kwargs.update(cmap=['blue'])
else:
kwargs.update(cmap=[color])
obj = datashade(hv_obj, **kwargs)
if spread:
obj = dynspread(obj)
return obj
def get_net(layout, bundled, shade):
if (layout.lower() == 'circle'):
net = hv.Graph.from_networkx(
G, nx.layout.circular_layout).relabel('Circular Layout').opts(
width=650, height=650, xaxis=None, yaxis=None, padding=0.1)
elif layout.lower() == 'spring':
net = hv.Graph.from_networkx(
G, nx.layout.spring_layout, k=0.8, iterations=100).relabel(
'Force-Directed Fruchterman-Reingold').opts(width=650,
height=650,
xaxis=None,
yaxis=None,
padding=0.1)
# , node_size=node_dict[i] for i in node_dict.keys())
else:
net = "Error 1: layout type must be: 'circle', 'spring'"
net.opts(width=650,
height=650,
xaxis=None,
yaxis=None,
padding=0.1,
edge_color_index='weight',
edge_cmap='jet')
if bundled:
net = bundle_graph(net)
if B_edge_select:
net.opts(inspection_policy='edges')
if shade:
net = hd.datashade(net).opts(plot=dict(height=650, width=650))
return net
def make_scatter(self, object_type, x_range=None, y_range=None, **kwargs):
self._set_data(**kwargs)
logging.info('x_range={}, y_range={}'.format(x_range, y_range))
if object_type == 'all':
dset = self.ds
else:
dset = self.ds.select(label=object_type)
pts = dset.to(hv.Points, kdims=['x', 'y'], vdims=['label'], groupby=[])
print(pts.dimensions())
scatter = dynspread(
datashade(pts,
x_range=x_range,
y_range=y_range,
dynamic=False,
normalization='log'))
hover = HoverTool(tooltips=[("(x,y)", "($x, $y)")])
# hv.opts({'RGB': {'plot' : {'tools' : [hover]}}}, scatter)
# scatter = scatter.opts(plot=dict(tools=[hover]))
title = '{} ({}) {}'.format(object_type, len(dset),
pts.get_dimension('y').label)
scatter = scatter.opts('RGB [width=600, height=400]').relabel(title)
return scatter
def spectrum_hires(wno, alb, legend=None, **kwargs):
"""Plot formated albedo spectrum
Parameters
----------
wno : float array, list of arrays
wavenumber
alb : float array, list of arrays
albedo
legend : list of str
legends for plotting
**kwargs : dict
Any key word argument for hv.opts
Returns
-------
bokeh plot
"""
import holoviews as hv
from holoviews.operation.datashader import datashade
hv.extension('bokeh')
kwargs['plot_height'] = kwargs.get('plot_height', 345)
kwargs['plot_width'] = kwargs.get('plot_width', 1000)
kwargs['y_axis_label'] = kwargs.get('y_axis_label', 'Albedo')
kwargs['x_axis_label'] = kwargs.get('x_axis_label', 'Wavelength [μm]')
kwargs['y_range'] = kwargs.get('y_range', [0, 1.2])
kwargs['x_range'] = kwargs.get('x_range', [0.3, 1])
points_og = datashade(hv.Curve((1e4 / wno, alb)))
return points_og
def hv_plot(G, layout='circular', method='bokeh', name='', kwargs=opts):
'''
Returns graph plot using HoloViews wrapper for bokeh.
Optionally, draws edges using datashade functions.
Accepted vars:
G => networkx.Graph() object
layout => circular; forceatlas2; random
method => bokeh; datashader
name => graph title or label
kwargs => optionals
'''
hv.extension("bokeh")
nodes, edges = nx_layout(G)
# apply parameters
if kwargs:
hv.opts.defaults(
hv.opts.EdgePaths(**kwargs),
hv.opts.Graph(**kwargs),
hv.opts.Nodes(**kwargs))
# plot edges with bokeh
circle = hv.Graph(edges, label=name).opts(style=dict(node_size=10))
# plot edges with datashader (WIP)
if method == 'datashader':
hnodes = circle.nodes.opts(style=dict(size=10))
dscirc = (hd.dynspread(hd.datashade(circle))*hnodes).relabel(name)
return dscirc
return circle
def make_view(self, x_range, y_range, **kwargs):
tiles = map_tiles.options(alpha=self.alpha, **opts1)
points = hv.Points(df, ['x', 'y'])
return tiles * datashade(points,
cmap=self.colormap,
x_range=x_range,
y_range=y_range,
**opts2)
def setUp(self):
from holoviews.tests.teststreams import Sum, Val
# kdims: a and b
self.dmap_ab = DynamicMap(lambda a, b: Points([a, b]),
kdims=['a', 'b']).redim.range(a=(0.0, 10.0),
b=(0.0, 10.0))
# kdims: b
self.dmap_b = DynamicMap(lambda b: Points([b, b]),
kdims=['b']).redim.range(b=(0.0, 10.0))
# no kdims, XY stream
self.xy_stream = XY()
self.dmap_xy = DynamicMap(lambda x, y: Points([x, y]),
streams=[self.xy_stream])
# no kdims, Z stream
self.z_stream = Z()
self.dmap_z = DynamicMap(lambda z: Points([z, z]),
streams=[self.z_stream])
# DynamicMap of a derived stream
self.stream_val1 = Val()
self.stream_val2 = Val()
self.stream_val3 = Val()
self.dmap_derived = DynamicMap(
lambda v: Points([v, v]),
streams=[
Sum([
self.stream_val1,
Sum([self.stream_val2, self.stream_val3])
])
])
if datashade is None:
return
# dmap produced by chained datashade and shade
self.px_stream = PX()
self.dmap_spread_points = spread(datashade(Points([0.0, 1.0])),
streams=[self.px_stream])
# data shaded with kdims: a, b
self.dmap_datashade_kdim_points = datashade(self.dmap_ab)
def view(self, cmap=None, shade=True):
raster_opts = dict(aggregator=self.aggregator, precompute=True)
cmap_opts = dict(cmap=cmap) if cmap else {}
if shade:
if cmap: raster_opts.update(cmap_opts)
shaded = datashade(self.obj, **raster_opts)
else:
shaded = rasterize(self.obj, **raster_opts).opts(style=cmap_opts)
return self.tiles * shaded * self.path + self.sections
def plot_datashade_dnb_figure(dataset, colormap, width=900, height=600):
return datashade(
hv.Scatter(dataset, kdims=["x"], vdims=['y', 'values']),
aggregator=ds.sum('values'),
cmap=colormap,
).opts(
width=width,
height=height,
)
def make_view(self, x_range=None, y_range=None, **kwargs):
#map_tiles = tiles.opts(style=dict(alpha=self.alpha), plot=options)
hvmap = projected_gv.select(z=self.altitude, k=self.trackrange)
dsmap = datashade(hvmap, x_sampling=1, y_sampling=1, cmap=self.colormap,
dynamic=False, x_range=x_range, y_range=y_range)
gv_options = {'bgcolor':'black', 'show_grid':True}
gvmap = dynspread(dsmap.opts(plot=gv_options))
return gvmap #* hv.util.Dynamic(aggregate(hvmap, width=5, height=5, streams=[PointerX]), operation=hv.QuadMesh)
def channel_map():
return dynspread(
datashade(
points, y_range=(0, 260),
streams=[xrange_stream])).opts(plot=dict(
width=600,
tools=[time_zoom, 'xpan'],
default_tools=['save', 'pan', 'box_zoom', 'save', 'reset'],
show_grid=False))
def view_elements(self, agg='any', line_color='black', cmap='black'):
""" Method to display the mesh as wireframe elements"""
if self.elements_toggle:
# return datashade(self.tri_mesh.edgepaths.opts(line_color=line_color), aggregator=agg,
# precompute=True, cmap=cmap)
return datashade(
self.tri_mesh.edgepaths.opts(
opts.TriMesh(edge_cmap='yellow', edge_color='yellow')))
else:
return hv.Curve([])
def gen_mesh(self, **kwargs):
if not self.draw_helper.poly_stream.element or self._clear:
self._clear = False
return RGB([], bounds=self.draw_helper.extent)
self._reset_mesh()
self.polys = self._process_polys(self.draw_helper.poly_stream.element)
self.refine_points = self.add_refine_points(self.draw_helper)
verts, tris = self.mesh.create_mesh()
return datashade(viz_mesh(verts, tris).edgepaths, cmap=['#000000'],
dynamic=False)
def draw_graph(G, label, cmap):
print(f'Shading {label}')
shaded = (datashade(G, normalization='linear', width=1000, height=1000) *
G.nodes).opts(
opts.Nodes(color=label,
width=1000,
cmap=cmap,
legend_position='right'))
hv.save(shaded, f'graphs/png/{label}.png')
hv.save(shaded, f'graphs/html/{label}.html')
def z_y_views(self, x_range, y_range):
x_min = x_range[0]; x_max = x_range[1]
y_min = y_range[0]; y_max = y_range[1]
datas = projected_gv.select(x=x_range, y=y_range, z=self.altitude, k=self.trackrange).to(hv.Dataset)
hvmap = datas.to(hv.Points, kdims=self.plotdims, vdims=['i', 'k'], groupby=[])
dsmap = datashade(hvmap, cmap=self.colormap, dynamic=False, x_range=x_range, y_range=y_range)
gv_options = {'bgcolor':'black', 'show_grid':True}
gvmap = dynspread(dsmap.opts(plot=gv_options))
return gvmap
def datashade(self, column=None, reduction=None):
"""Return datashader shaded with reduction of given column value
Arguments:
- `column`:
- `reduction`:
"""
import datashader as ds
from bokeh import palettes
from holoviews.operation.datashader import aggregate, shade, datashade, dynspread
if reduction is None:
reduction = ds.mean
#print "datashade", column, reduction
shade_opts = {"cmap": palettes.inferno(64)}
if column is not None:
d = self._points.dframe()[column]
d_min, d_max = d.min(), d.max()
#print "Value Range:", d_min, d_max
#shade_opts["clims"] = (d_min, d_max)
if d_min * d_max < 0.0:
print("Diverging palette")
d_extreme = max(-d_min, d_max)
shade_opts = {
"cmap": palettes.RdYlBu11,
#"clims": (-d_extreme, d_extreme)
}
def _linear_norm_debug(v, masked):
import pandas as pd
#print args
#print kwargs
print(v)
print(pd.DataFrame(v).describe())
min_v, max_v = v[~masked].min(), v[~masked].max()
print(min_v, max_v)
o = (v - min_v) / (max_v - min_v)
print(o)
return o
#del (shade_opts["clims"])
plot = dynspread(
datashade(
self._points,
aggregator=ds.count() if column is None else reduction(column),
normalization="linear" if "clims" in shade_opts else "eq_hist",
**shade_opts))
return plot
import dask.dataframe as dd
import holoviews as hv
import geoviews as gv
from bokeh.models import WMTSTileSource
from holoviews.operation.datashader import datashade
url = 'https://server.arcgisonline.com/ArcGIS/rest/services/World_Imagery/MapServer/tile/{Z}/{Y}/{X}.jpg'
wmts = WMTSTileSource(url=url)
# 1. Instead of using hv.extension, declare that we are using Bokeh and grab the renderer
renderer = hv.renderer('bokeh')
# 2. Declare points and datashade them
ddf = dd.read_csv('../data/nyc_taxi.csv', usecols=['pickup_x', 'pickup_y']).persist()
shaded = datashade(hv.Points(ddf))
# Set some plot options
app = gv.WMTS(wmts) * shaded.opts(plot=dict(width=800, height=600))
# 3. Instead of Jupyter's automatic rich display, render the object as a bokeh document
doc = renderer.server_doc(app)
doc.title = 'HoloViews Bokeh App'
import dask.dataframe as dd
from holoviews.operation.datashader import datashade, aggregate, shade
from bokeh.models import WMTSTileSource
renderer = hv.renderer('bokeh')
# Load data
usecols = ['tpep_pickup_datetime', 'dropoff_x', 'dropoff_y']
ddf = dd.read_csv('../data/nyc_taxi.csv', parse_dates=['tpep_pickup_datetime'], usecols=usecols)
ddf['hour'] = ddf.tpep_pickup_datetime.dt.hour
ddf = ddf.persist()
# Declare objects
stream = hv.streams.Counter()
points = hv.Points(ddf, kdims=['dropoff_x', 'dropoff_y'])
dmap = hv.DynamicMap(lambda counter: points.select(hour=counter%24).relabel('Hour: %s' % (counter % 24)),
streams=[stream])
shaded = datashade(dmap)
hv.opts('RGB [width=800, height=600, xaxis=None, yaxis=None]')
url = 'https://server.arcgisonline.com/ArcGIS/rest/services/World_Imagery/MapServer/tile/{Z}/{Y}/{X}.jpg'
wmts = gv.WMTS(WMTSTileSource(url=url))
overlay = wmts * shaded
# Create server document
doc = renderer.server_doc(overlay)
dmap.periodic(1)
import dask.dataframe as dd
import holoviews as hv
from holoviews.operation.datashader import datashade
# 1. Instead of using hv.extension, declare that we are using Bokeh and grab the renderer
renderer = hv.renderer('bokeh')
# 2. Load data and datashade it
ddf = dd.read_csv('../data/nyc_taxi.csv', usecols=['dropoff_x', 'dropoff_y']).persist()
points = hv.Points(ddf, kdims=['dropoff_x', 'dropoff_y'])
shaded = datashade(points).opts(plot=dict(width=800, height=600))
# 3. Instead of Jupyter's automatic rich display, render the object as a bokeh document
doc = renderer.server_doc(shaded)
doc.title = 'HoloViews Bokeh App'
