bokeh.models.Range1d

    def _initialize_limits(self, p):
        if self.xlim is not None:
            p.x_range = models.Range1d(self.xlim[0], self.xlim[-1])
        if self.ylim is not None:
            p.y_range = models.Range1d(self.ylim[0], self.ylim[-1])

    @staticmethod

def patches(x, y, diff):
    y = diff.set_index(x)[y]
    range = y.abs().max() * 1.1
    green = y.clip(lower=0).append(pd.Series([0, 0], index=[y.index[len(y) - 1], y.index[0]]))
    red = y.clip(upper=0).append(pd.Series([0, 0], index=[y.index[len(y) - 1], y.index[0]]))
    return green, red, Range1d(start=-range, end=range)


def add_tsid_line(f, x, y, source, color='black', line_dash='solid'):

def add_cum_line(f, y, source, color='black', line_dash='solid'):
    y_c = source[y].sort_values(ascending=False).reset_index(drop=True)
    f.line(x=y_c.index, y=y_c, line_color=color, line_dash=line_dash)
    f.x_range = Range1d(start=y_c.index.max(), end=y_c.index.min())
    df = y_c.to_frame('y')
    df['x'] = df.index
    return df


def cumulative_plot(nx, ny, y, source, other=None, ylo=None, yhi=None, output_backend=DEFAULT_BACKEND):

def histogram_plot(nx, ny, x, source, xlo=None, xhi=None, nsub=5, output_backend=DEFAULT_BACKEND):
    if not present(source, x): return None
    xlo, xhi = source[x].min() if xlo is None else xlo, source[x].max() if xhi is None else xhi
    bins = pd.interval_range(start=xlo, end=xhi, periods=nsub * (xhi - xlo), closed='left')
    c = [palettes.Inferno[int(xhi - xlo + 1)][int(b.left - xlo)] for b in bins]
    hrz_categorized = pd.cut(source[x], bins)
    counts = hrz_categorized.groupby(hrz_categorized).count()
    f = figure(output_backend=output_backend, plot_width=nx, plot_height=ny, x_range=Range1d(start=xlo, end=xhi),
               x_axis_label=x)
    f.quad(left=counts.index.categories.left, right=counts.index.categories.right, top=counts, bottom=0,
           color=c, fill_alpha=0.2)
    f.toolbar_location = None
    f.yaxis.visible = False
    return f


def add_route(f, source, color='black', line_dash='solid'):

def make_range(column, lo=None, hi=None):
    column = column.dropna()
    mn, mx = column.min() if lo is None else lo, column.max() if hi is None else hi
    delta = mx - mn
    return Range1d(start=mn - 0.1 * delta, end=mx + 0.1 * delta)


def evenly_spaced_hues(n, saturation=1, value=1, stagger=1):

def render_network_bokeh(CRN,layout="force", layoutfunc = None, plot_reactions = True, plot_species = True, plot_edges = True, plot_arrows = True,
                        iterations=2000,rseed=30,posscale=1,export=False,species_glyph_size = 12, reaction_glyph_size = 8, export_name = None, **keywords):
    DG, DGspec, DGrxn = generate_networkx_graph(CRN,**keywords) #this creates the networkx objects
    plot = Plot(plot_width=500, plot_height=500, x_range=Range1d(-500, 500), y_range=Range1d(-500, 500)) #this generates a 
    show_im = False
    images = None
    if("imagedict" in keywords and keywords["imagedict"] is not None):
        show_im = True
    if(export):
        plot.output_backend = "svg"
    graphPlot(DG,DGspec,DGrxn,plot,layout=layout,posscale=posscale,iterations=iterations,rseed=rseed,show_species_images=show_im,
        layoutfunc=layoutfunc, plot_reactions = plot_reactions, plot_species = plot_species, plot_edges = plot_edges, plot_arrows = plot_arrows,
        species_glyph_size = species_glyph_size, reaction_glyph_size = reaction_glyph_size) 
    if(export):
        if export_name is None:
            raise ValueError("To export you must supply export_name keyword.")
        export_svgs(plot,filename=export_name+".svg")
    return plot

def getSuspFigure(df):
    vars = ['speedkmh', 'sus_travel_lf', 'sus_travel_lr',
            'sus_travel_rf', 'sus_travel_rr']
    tools = ['time', 'dist']+vars
    return getSimpleFigure(df, vars, tools,
                           {"sus_travel": Range1d(start=-10, end=120)})

def getBrakeTempFigure(df):

def getBrakeTempFigure(df):
    vars = ['brake_temp_lf','brake_temp_lr',
            'brake_temp_rf', 'brake_temp_rr',
            'throttle','brake']
    tools = ['time', 'dist', 'speed']+vars
    return getSimpleFigure(df, vars+['tc', 'abs'], tools,
                           {"pedals": Range1d(start=-20, end=700), "tcabs": Range1d(start=-1, end=20)},
                           {"pedals":['throttle','brake'], "tcabs":['tc', 'abs']})

def getWheelSpeedFigure(df):

def getWheelSpeedFigure(df):
    vars = ['wheel_speed_lf','wheel_speed_lr',
            'wheel_speed_rf', 'wheel_speed_rr',
            'throttle','brake']
    tools = ['time','dist','speed']+vars
    return getSimpleFigure(df, vars+['tc', 'abs'], tools,
                           {"pedals": Range1d(start=-20, end=400), "tcabs": Range1d(start=-1, end=20)},
                           {"pedals":['throttle','brake'], "tcabs":['tc', 'abs']})

def getTyreTairFigure(df):

def getTyreTairFigure(df):
    vars = ['tyre_tair_lf','tyre_tair_lr',
            'tyre_tair_rf', 'tyre_tair_rr',
            'throttle','brake']
    tools = ['time', 'dist', 'speed']+vars
    return getSimpleFigure(df, vars+['tc', 'abs'], tools,
                           {"pedals": Range1d(start=-20, end=400), "tcabs": Range1d(start=-1, end=20)},
                           {"pedals":['throttle','brake'], "tcabs":['tc', 'abs']})

def getTyrePreassureFigure(df):

def getTyrePreassureFigure(df):
    vars = ['speedkmh', 'tyre_press_lf','tyre_press_lr',
            'tyre_press_rf', 'tyre_press_rr']
    tools = ['time', 'dist']+vars
    return getSimpleFigure(df, vars, tools,
                           {"tyre_press": Range1d(start=26, end=32)})

def getOversteerFigure(df):

def getOversteerFigure(df):
    vars = ['speedkmh','oversteer','understeer']#,'steering_corr','neutral_steering']
    tools = ['time','dist']+vars
    p0 = getSimpleFigure(df, vars+['tc', 'throttle','brake'], tools,
                           {"pedals": Range1d(start=-10, end=500), "tc": Range1d(start=-1, end=50), "oversteer": Range1d(start=-15, end=25)},
                           {"pedals":['throttle','brake'], "tc":['tc','g_lat'], "oversteer":['steering_corr','neutral_steering','oversteer','understeer']})

    vars = ['g_lat', 'g_lon', 'g_sum','steering_corr','neutral_steering', 'oversteer', 'understeer']
    tools = ['time','dist']+vars
    p1 = getSimpleFigure(df, vars, tools,
                          {"oversteer": Range1d(start=-15, end=35)},
                          {"oversteer":['steering_corr','neutral_steering',
                                        'oversteer','understeer']},
                          x_range = p0.x_range)

    return gridplot([p0, p1], ncols=1, sizing_mode='scale_width')


def getLapDelta():

    def test_model(self):
        m = Range1d(start=10, end=20)
        assert self.encoder.default(m) == m.ref
        assert isinstance(self.encoder.default(m), dict)

    def test_hasprops(self):

    def test_basic(self):
        r = Range1d()
        check_properties_existence(r, [
            "callback",
            "start",
            "end",
            "reset_start",
            "reset_end",
            "bounds",
            "min_interval",
            "max_interval"],
        )

    def test_init_with_timedelta(self):

    def test_init_with_timedelta(self):
        range1d = Range1d(start=-dt.timedelta(seconds=5), end=dt.timedelta(seconds=3))
        assert range1d.start == -dt.timedelta(seconds=5)
        assert range1d.end == dt.timedelta(seconds=3)
        assert range1d.bounds is None

    def test_init_with_datetime(self):

    def test_init_with_datetime(self):
        range1d = Range1d(start=dt.datetime(2016, 4, 28, 2, 20, 50), end=dt.datetime(2017, 4, 28, 2, 20, 50))
        assert range1d.start == dt.datetime(2016, 4, 28, 2, 20, 50)
        assert range1d.end == dt.datetime(2017, 4, 28, 2, 20, 50)
        assert range1d.bounds is None

    def test_init_with_float(self):

    def test_init_with_float(self):
        range1d = Range1d(start=-1.0, end=3.0)
        assert range1d.start == -1.0
        assert range1d.end == 3.0
        assert range1d.bounds is None

    def test_init_with_int(self):

    def test_init_with_int(self):
        range1d = Range1d(start=-1, end=3)
        assert range1d.start == -1
        assert range1d.end == 3
        assert range1d.bounds is None

    def test_init_with_positional_arguments(self):

    def test_init_with_positional_arguments(self):
        range1d = Range1d(1, 2)
        assert range1d.start == 1
        assert range1d.end == 2
        assert range1d.bounds is None

    def test_init_with_keyword_arguments(self):

    def test_init_with_keyword_arguments(self):
        range1d = Range1d(start=1, end=2)
        assert range1d.start == 1
        assert range1d.end == 2
        assert range1d.bounds is None

    def test_cannot_initialize_with_both_keyword_and_positional_arguments(self):

    def test_cannot_initialize_with_both_keyword_and_positional_arguments(self):
        with pytest.raises(ValueError):
            Range1d(1, 2, start=1, end=2)


    def test_cannot_initialize_with_three_positional_arguments(self):

    def test_cannot_initialize_with_three_positional_arguments(self):
        with pytest.raises(ValueError):
            Range1d(1, 2, 3)


    def test_with_max_bound_smaller_than_min_bounded_raises_valueerror(self):

    def test_with_max_bound_smaller_than_min_bounded_raises_valueerror(self):
        with pytest.raises(ValueError):
            Range1d(1, 2, bounds=(1, 0))
        with pytest.raises(ValueError):
            Range1d(1, 2, bounds=[1, 0])


    def test_bounds_with_text_rejected_as_the_correct_value_error(self):

    def test_bounds_with_text_rejected_as_the_correct_value_error(self):
        with pytest.raises(ValueError) as e:
            Range1d(1, 2, bounds="21")  # The string is indexable, so this may not fail properly
        assert e.value.args[0].startswith('expected an element of either')


    def test_bounds_with_three_item_tuple_raises_valueerror(self):

    def test_init_with_bad_bounds(self):
        with pytest.raises(ValueError):
            DataRange1d(1, 2, bounds=(1, 0))
        with pytest.raises(ValueError):
            DataRange1d(1, 2, bounds=[1, 0])
        with pytest.raises(ValueError):
            Range1d(1, 2, bounds="21")


class Test_FactorRange(object):

    def __init__(self, df, scale=13, border_day=0.25, border_month=0.4, border_year=0.4, title=None,
                 hover=None, not_hover=tuple(), all_hover=False):

        self._hover_args = (hover, not_hover, all_hover)

        start, finish = df.index.min(), df.index.max()
        start = dt.date(start.year, 1, 1)
        finish = dt.date(finish.year + 1, 1, 1) - dt.timedelta(days=1)
        delta_year = 7 * (1 + border_day) + border_year

        self._df = df
        self._df_all = pd.DataFrame(data={CALENDAR_SIZE: 1, CALENDAR_RADIUS: 0.5},
                                    index=pd.date_range(start, finish))

        self._set_date(self._df)
        self._set_xy(self._df, start, finish, border_day, border_month, delta_year)
        self._set_date(self._df_all)
        self._set_xy(self._df_all, start, finish, border_day, border_month, delta_year)

        xlo = self._df_all[CALENDAR_X].min()
        xhi = self._df_all[CALENDAR_X].max()
        ylo = self._df_all[CALENDAR_Y].min()
        yhi = self._df_all[CALENDAR_Y].max()
        nx = int(scale * (xhi - xlo))
        ny = int(scale * (yhi - ylo))

        plot = Plot(x_range=Range1d(xlo - 4, xhi + 4), y_range=Range1d(ylo - 1, yhi + 4),
                    width=nx, height=ny, title=Title(text=title))
        self._add_labels(plot, start, finish, xlo, xhi, yhi, border_day, delta_year)
        self._plot = plot

    @staticmethod

def circuit_from(bqm):
    G = bqm.to_networkx_graph()
    plot = Plot(
        plot_width=600, plot_height=400, x_range=Range1d(-0.1, 1.1), y_range=Range1d(-0.1, 1.1)
    )
    plot.title.text = "Multiplication as a BQM"
    plot.add_tools(HoverTool(tooltips=None), TapTool(), BoxSelectTool())
    graph_renderer = from_networkx(G, circuit_layout)

    circle_size = 25
    graph_renderer.node_renderer.glyph = Circle(size=circle_size, fill_color="#F5F7FB")
    graph_renderer.node_renderer.selection_glyph = Circle(size=circle_size, fill_color="#EEA64E")
    graph_renderer.node_renderer.hover_glyph = Circle(size=circle_size, fill_color="#FFE86C")

    edge_size = 2
    graph_renderer.edge_renderer.glyph = MultiLine(
        line_color="#CCCCCC", line_alpha=0.8, line_width=edge_size
    )
    graph_renderer.edge_renderer.selection_glyph = MultiLine(
        line_color="#EEA64E", line_width=edge_size
    )
    graph_renderer.edge_renderer.hover_glyph = MultiLine(line_color="#FFE86C", line_width=edge_size)

    graph_renderer.selection_policy = NodesAndLinkedEdges()
    graph_renderer.inspection_policy = NodesAndLinkedEdges()

    plot.renderers.append(graph_renderer)

    plot.background_fill_color = "#202239"

    add_labels(plot)
    show(Row(plot))


def frequency_of(results):

def draw(S, position=None, with_labels=False):
    """Plot the given signed social network.

    Args:
        S: The network
        position (dict, optional):
            The position for the nodes. If no position is provided, a layout will be calculated. If the nodes have
            'color' attributes, a Kamanda-Kawai layout will be used to group nodes of the same color together.
            Otherwise, a circular layout will be used.

    Returns:
        A dictionary of positions keyed by node.

    Examples:
    >>> import dwave_structural_imbalance_demo as sbdemo
    >>> gssn = sbdemo.GlobalSignedSocialNetwork()
    >>> nld_before = gssn.get_node_link_data('Syria', 2013)
    >>> nld_after = gssn.solve_structural_imbalance('Syria', 2013)
    # draw Global graph before solving; save node layout for reuse
    >>> position = sbdemo.draw('syria.png', nld_before)
    # draw the Global graph; reusing the above layout, and calculating a new grouped layout
    >>> sbdemo.draw('syria_imbalance.png', nld_after, position)
    >>> sbdemo.draw('syria_imbalance_grouped', nld_after)

    """

    # we need a consistent ordering of the edges
    edgelist = S.edges()
    nodelist = S.nodes()

    def layout_wrapper(S):
        pos = position
        if pos is None:
            try:
                # group bipartition if nodes are colored
                dist = defaultdict(dict)
                for u, v in product(nodelist, repeat=2):
                    if u == v:  # node has no distance from itself
                        dist[u][v] = 0
                    elif (
                        nodelist[u]["color"] == nodelist[v]["color"]
                    ):  # make same color nodes closer together
                        dist[u][v] = 1
                    else:  # make different color nodes further apart
                        dist[u][v] = 2
                pos = nx.kamada_kawai_layout(S, dist)
            except KeyError:
                # default to circular layout if nodes aren't colored
                pos = nx.circular_layout(S)
        return pos

    # call layout wrapper once with all nodes to store position for calls with partial graph
    position = layout_wrapper(S)

    plot = Plot(
        plot_width=600, plot_height=400, x_range=Range1d(-1.2, 1.2), y_range=Range1d(-1.2, 1.2)
    )
    tools = [WheelZoomTool(), ZoomInTool(), ZoomOutTool(), PanTool()]
    plot.add_tools(*tools)
    plot.toolbar.active_scroll = tools[0]

    def get_graph_renderer(S, line_dash):
        # we need a consistent ordering of the edges
        edgelist = S.edges()
        nodelist = S.nodes()

        # get the colors assigned to each edge based on friendly/hostile
        sign_edge_color = ["#87DACD" if S[u][v]["sign"] == 1 else "#FC9291" for u, v in edgelist]

        # get the colors assigned to each node by coloring
        try:
            coloring_node_color = [
                "#4378F8" if nodelist[v]["color"] else "#FFE897" for v in nodelist
            ]
        except KeyError:
            coloring_node_color = ["#FFFFFF" for __ in nodelist]

        graph_renderer = from_networkx(S, layout_wrapper)

        circle_size = 10
        graph_renderer.node_renderer.data_source.add(coloring_node_color, "color")
        graph_renderer.node_renderer.glyph = Circle(size=circle_size, fill_color="color")

        edge_size = 2
        graph_renderer.edge_renderer.data_source.add(sign_edge_color, "color")
        try:
            graph_renderer.edge_renderer.data_source.add(
                [S[u][v]["event_year"] for u, v in edgelist], "event_year"
            )
            graph_renderer.edge_renderer.data_source.add(
                [S[u][v]["event_description"] for u, v in edgelist], "event_description"
            )
            plot.add_tools(
                HoverTool(
                    tooltips=[("Year", "@event_year"), ("Description", "@event_description")],
                    line_policy="interp",
                )
            )
        except KeyError:
            pass
        graph_renderer.edge_renderer.glyph = MultiLine(line_color="color", line_dash=line_dash)

        graph_renderer.inspection_policy = EdgesAndLinkedNodes()

        return graph_renderer

    try:
        S_dash = S.edge_subgraph(((u, v) for u, v in edgelist if S[u][v]["frustrated"]))
        S_solid = S.edge_subgraph(((u, v) for u, v in edgelist if not S[u][v]["frustrated"]))
        plot.renderers.append(get_graph_renderer(S_dash, "dashed"))
        plot.renderers.append(get_graph_renderer(S_solid, "solid"))
    except KeyError:
        plot.renderers.append(get_graph_renderer(S, "solid"))

    plot.background_fill_color = "#202239"

    positions = layout_wrapper(S)
    if with_labels:
        data = {"xpos": [], "ypos": [], "label": []}
        for label, pos in positions.items():
            data["label"].append(label)
            data["xpos"].append(pos[0])
            data["ypos"].append(pos[1])

        labels = LabelSet(
            x="xpos",
            y="ypos",
            text="label",
            level="glyph",
            source=ColumnDataSource(data),
            x_offset=-5,
            y_offset=10,
            text_color="#F5F7FB",
            text_font_size="12pt",
        )
        plot.add_layout(labels)

    show(Row(plot))

    return positions

def get_bokeh(df, linear = False):
    
    #df = df.fillna("")

    X=0
    Y=0

    TOOLTIPS='  <  font size="3"> < b>@Feature < /b> — @Type   @pi_permatch_int < /font>  < br> @Description'

    hover = HoverTool(names=["features"])
    plotSize=.35
    plotDimen=800

    x_range = Range1d(-plotSize, plotSize, bounds=(-.5, .5), min_interval=.1)
    y_range = Range1d(-plotSize, plotSize, bounds=(-.5, .5), min_interval=.1)
    toolbar = None
    p = figure(plot_height=plotDimen,plot_width=plotDimen, title="",
                toolbar_location=toolbar, toolbar_sticky=False, match_aspect=True,
                sizing_mode='scale_width', tools=['save',hover,'pan'], tooltips=TOOLTIPS,
                #x_range=(-plotSize, plotSize), y_range=(-plotSize, plotSize))
                x_range=x_range, y_range=y_range)
    p.toolbar.logo = None
    p.add_tools(WheelZoomTool(zoom_on_axis=False))
    p.toolbar.active_scroll = p.select_one(WheelZoomTool)

    #backbone line
    p.circle(x=X, y=Y, radius=baseRadius, line_color="#000000", fill_color=None, line_width=2.5)

    df = calc_level(df)
    
    if linear:
        line_length = baseRadius / 5
        p.line([0,0], [baseRadius - line_length, baseRadius + line_length],
                line_width = 4, level="overlay", line_color = "black")

    df['pi_permatch_int']=df['pi_permatch'].astype('int')
    df['pi_permatch_int'] = df['pi_permatch_int'].astype(str) + "%"
    df.loc[df['db'] == "Rfam", 'pi_permatch_int'] = "" #removes percent from infernal hits

    df['rstart']=((df["qstart"]/df["qlen"])*2*pi)
    df['rend']  =((df["qend"]/df["qlen"])*2*pi)
    df['rstart'] = np.where(df['rstart']  <  0, df['rstart'] + (2*pi), df['rstart'])
    df['rend'] = np.where(df['rend']  <  0, df['rend'] + (2*pi), df['rend'])
    df['rend'] = np.where(df['rend']  <  df['rstart'], df['rend'] + (2*pi), df['rend'])

    df['Type'] = df['Type'].str.replace('rep_origin','origin of replication')
    
    #DDE0BD
    #C97064
    #C9E4CA
    fullColorDf=pd.read_csv(rsc.get_resource("data", "colors.csv"),index_col=0)
    fragColorDf=fullColorDf.copy()
    fragColorDf[['fill_color','line_color']]=fragColorDf[['line_color','fill_color']]
    fragColorDf["fill_color"]="#ffffff"

    full=df[df["fragment"]==False]
    full=full.merge(fullColorDf,how = "left",on=["Type"])
    full['legend'] = full['Type']
    full=full.fillna({"color":"grey","fill_color":"#808080","line_color":"#000000"})

    frag=df[df["fragment"]==True]
    frag=frag.merge(fragColorDf,how = "left",on=["Type"])
    frag=frag.fillna({"color":"grey","fill_color":"#ffffff","line_color":"#808080"})

    df=full.append(frag).reset_index(drop=True)#.set_index('Feature')

    # add orientation column
    orient = pd.read_csv(rsc.get_resource("data", "feature_orientation.csv"),header=None, names = ["Type","has_orientation"])
    orient['Type'] = orient['Type']
    orient['has_orientation'] = orient['has_orientation'].map({"T":True})
    df = df.merge(orient, on="Type", how = "left")
    df['Type'] = df['Type'].str.replace("_"," ")
    df['has_orientation'] = df['has_orientation'].fillna(value=False)

    df[['x','y',"Lx1","Ly1","annoLineColor","lineX","lineY","theta","text_align"]]=df.apply(calc_glyphs,axis=1)

    df['legend'] = df['Type']
    #allowedTypes = ['CDS',"promoter","origin of replication","swissprot"]
    allowedTypes = fullColorDf['Type']
    mask = ~df['legend'].isin(allowedTypes)
    df.loc[mask, 'legend'] = 'misc feature'

    #plot annotations
    source = ColumnDataSource(df)
    p.patches('x', 'y', fill_color='fill_color', line_color='line_color',
            name="features", line_width=2.5, source=source, legend_group = "legend")
    p.multi_line(xs="lineX", ys="lineY", line_color="annoLineColor", line_width=3,
            level="overlay",line_cap='round',alpha=.5, source=source)

    #`text_align` cannot read from `source` -- have to do this workaround
    right=ColumnDataSource(df[df['text_align']=='right'])
    left=ColumnDataSource(df[df['text_align']=='left'])
    bCenter=ColumnDataSource(df[df['text_align']=='b_center'])
    tCenter=ColumnDataSource(df[df['text_align']=='t_center'])

    text_level = 'overlay'
    p.text(x="Lx1", y="Ly1",name="2",x_offset=3,y_offset=8, text_align="left",
            text='Feature', level=text_level, source=right)
    p.text(x="Lx1", y="Ly1",name="2",x_offset=-5,y_offset=8, text_align="right",
            text='Feature', level=text_level, source=left)
    p.text(x="Lx1", y="Ly1",name="2",x_offset=0,y_offset=15, text_align="center",
            text='Feature', level=text_level, source=bCenter)
    p.text(x="Lx1", y="Ly1",name="2",x_offset=0,y_offset=0, text_align="center",
            text='Feature', level=text_level, source=tCenter)

    #calculate chunk size(s) for drawing lines
    plasLen = df.iloc[0]['qlen']
    ticks = calc_num_markers(plasLen)
    ticks_cds = ColumnDataSource(ticks)
    p.multi_line(xs="lineX", ys="lineY", line_color="black", line_width=2,
                level="underlay", line_cap='round', alpha = .5, source = ticks_cds)

    right=ColumnDataSource(ticks[ticks['text_align']=='right'])
    left=ColumnDataSource(ticks[ticks['text_align']=='left'])
    bCenter=ColumnDataSource(ticks[ticks['text_align']=='b_center'])
    tCenter=ColumnDataSource(ticks[ticks['text_align']=='t_center'])
    p.text(x="Lx1", y="Ly1",name="2",x_offset=3,y_offset=6, text_align="left",
            text='bp', alpha = .5, text_font_size = 'size',level=text_level, source=right)
    p.text(x="Lx1", y="Ly1",name="2",x_offset=-5,y_offset=6, text_align="right",
            text='bp', alpha = .5, text_font_size = 'size',level=text_level, source=left)
    p.text(x="Lx1", y="Ly1",name="2",x_offset=0,y_offset=15, text_align="center",
            text='bp', alpha = .5, text_font_size = 'size',level=text_level, source=bCenter)
    p.text(x="Lx1", y="Ly1",name="2",x_offset=0,y_offset=-3, text_align="center",
            text='bp', alpha = .5, text_font_size = 'size', level=text_level, source=tCenter)

    p.add_layout(Label(x=0, y=0,name="2",x_offset=0,y_offset=-8, text_align="center",
            text=f"{plasLen} bp", text_color = "#7b7b7b", text_font_size = '16px', level=text_level))

    p.axis.axis_label=None
    p.axis.visible=False
    p.grid.grid_line_color = "#EFEFEF"
    p.outline_line_color = "#DDDDDD"
    p.legend.location = 'bottom_left'
    p.legend.border_line_color = "#EFEFEF"
    p.legend.visible = True

    #st.write(df)

    return p

def graphPlot(DG,DGspecies,DGreactions,plot,layout="force",positions=None,plot_species = True, plot_reactions = True, plot_edges = True, plot_arrows = True,\
              species_glyph_size = 12, reaction_glyph_size = 8, posscale = 1.0,layoutfunc=None,iterations=2000,rseed=30,show_species_images=False):
    """given a directed graph, plot it!
    Inputs:
    DG: a directed graph of type DiGraph
    DGspecies: a directed graph which only contains the species nodes
    DGreactions: a directed graph which only contains the reaction nodes
    plot: a bokeh plot object
    layout: graph layout function. 
                'force' uses fa2 to push nodes apart
                'circle' plots the nodes and reactions in two overlapping circles, with the reactions on the inside of the circle
                'custom' allows user input "layoutfunc". Internally, layoutfunc is passed the three inputs (DG, DGspecies, DGreactions)
                                                        and should output a position dictionary with node {  <  node number>:(x,y)}

    positions: a dictionary of node names and x,y positions. this gets passed into the layout function
    posscale: multiply the scaling of the plot. This only affects the arrows because the arrows are a hack :("""
    random.seed(rseed)
    if(not PLOT_NETWORK):
        warn("network plotting disabled because some libraries are not found")
        return
    if(layout == "force"):
        # below are parameters for the force directed graph visualization
        forceatlas2 = ForceAtlas2(
            # Behavior alternatives
            outboundAttractionDistribution=True,  # Dissuade hubs
            linLogMode=False,  # NOT IMPLEMENTED
            # Prevent overlap (NOT IMPLEMENTED)
            adjustSizes=False,
            edgeWeightInfluence=1.0,

            # Performance
            jitterTolerance=1.0,  # Tolerance
            barnesHutOptimize=True,
            barnesHutTheta=1.2,
            multiThreaded=False,  # NOT IMPLEMENTED

            # Tuning
            scalingRatio=2.4*posscale,
            strongGravityMode=False,
            gravity=1.0,

            # Log
            verbose=False)

        positions = forceatlas2.forceatlas2_networkx_layout(
            DG, pos=positions, iterations=iterations)
    elif(layout == "circle"):
        positions = nx.circular_layout(DGspecies, scale=50*posscale)
        positions.update(nx.circular_layout(DGreactions, scale=35*posscale))
    elif(layout == "custom"):
        positions = layoutfunc(DG, DGspecies, DGreactions)
    reaction_renderer = from_networkx(DGreactions, positions, center=(0, 0))
    species_renderer = from_networkx(DGspecies, positions, center=(0, 0))
    edges_renderer = from_networkx(DG, positions, center=(0, 0))

    #Set xbounds and ybounds:
    xbounds = [0, 0]
    ybounds = [0, 0]
    for n in positions:
        xbounds[0] = min([xbounds[0], positions[n][0]])
        xbounds[1] = max([xbounds[1], positions[n][0]])
        ybounds[0] = min([ybounds[0], positions[n][1]])
        ybounds[1] = max([ybounds[1], positions[n][1]])

    max_glyph = max([reaction_glyph_size, species_glyph_size])
    xbounds[0] -= max_glyph
    xbounds[1] += max_glyph
    ybounds[0] -= max_glyph
    ybounds[1] += max_glyph

    # edges
    edges_renderer.node_renderer.glyph = Circle(
        size=species_glyph_size, line_alpha=0, fill_alpha=0, fill_color="color")
    edges_renderer.edge_renderer.glyph = MultiLine(
        line_alpha=0.2, line_width=4, line_join="round", line_color="color")
    edges_renderer.edge_renderer.selection_glyph = MultiLine(
        line_color=Spectral4[2], line_width=5, line_join="round")
    edges_renderer.edge_renderer.hover_glyph = MultiLine(
        line_color=Spectral4[1], line_width=5, line_join="round")
    if plot_arrows:
        xbounds_a, ybounds_a = makeArrows2(
            edges_renderer, DG, positions, headsize=5)  # make the arrows!
        xbounds[0] = min([xbounds[0], xbounds_a[0]])
        xbounds[1] = max([xbounds[1], xbounds_a[1]])
        ybounds[0] = min([ybounds[0], ybounds_a[0]])
        ybounds[1] = max([ybounds[1], ybounds_a[1]])

    # we want to find the middle of the graph and plot a square that is 1:1 aspect ratio

    # find the midpoint of the graph
    xmid = statistics.mean(xbounds)
    ymid = statistics.mean(ybounds)
    # now, subtract the middle from the edges
    xmiddlized = [a-xmid for a in xbounds]
    ymiddlized = [a-ymid for a in ybounds]
    # now, find the biggest dimension
    absdim = max([abs(a) for a in xmiddlized+ymiddlized])
    xlim = [xmid-absdim*1.05, xmid + absdim*1.05]
    ylim = [ymid-absdim*1.05, ymid + absdim*1.05]
    # now set it on the plot!
    plot.x_range = Range1d(xlim[0], xlim[1])
    plot.y_range = Range1d(ylim[0], ylim[1])

    # reactions
    reaction_renderer.node_renderer.glyph = Square(
        size=reaction_glyph_size, fill_color="color")
    reaction_renderer.node_renderer.selection_glyph = Square(
        size=reaction_glyph_size, fill_color=Spectral4[2])
    reaction_renderer.node_renderer.hover_glyph = Square(
        size=reaction_glyph_size, fill_color=Spectral4[1])

    # nodes
    species_renderer.node_renderer.glyph = Circle(size=12, fill_color="color")
    species_renderer.node_renderer.selection_glyph = Circle(size=15, fill_color=Spectral4[2])
    species_renderer.node_renderer.hover_glyph = Circle(size=15, fill_color=Spectral4[1])
    
    #this part adds the interactive elements that make it so that the lines are highlighted 
    #when you mouse over and click
    edge_hover_tool = HoverTool(tooltips= None,renderers=[edges_renderer])
    if( not show_species_images):
        species_hover_tool = HoverTool(tooltips=[("name", "@species"), ("type", "@type")],\
                                        renderers=[species_renderer],attachment="right")
    else:
        species_hover_tool = HoverTool(tooltips=' < div> < div>  < img src="data:image/png;base64,@image" style="float: left; margin: 0px 0px 0px 0px;"> < /img> < /div> < /div>',\
                                        renderers=[species_renderer],attachment="right")
    rxn_hover_tool = HoverTool(tooltips=[("reaction", "@species"), ("type", "@type"),("k_f","@k"),("k_r","@k_r")],\
                                        renderers=[reaction_renderer],attachment="right")
    
    plot.add_tools(edge_hover_tool,species_hover_tool,rxn_hover_tool, TapTool(), BoxSelectTool(),PanTool(),WheelZoomTool())

    edges_renderer.selection_policy = NodesAndLinkedEdges()
    edges_renderer.inspection_policy = EdgesAndLinkedNodes()

    if plot_edges:
        plot.renderers.append(edges_renderer)
    if plot_reactions:
        plot.renderers.append(reaction_renderer)
    if plot_species:
        plot.renderers.append(species_renderer)

def generate_networkx_graph(CRN,useweights=False,use_pretty_print=False,pp_show_material=True,

    def plot(self, pm: PrettyMIDI):
        """
        Plots the pretty midi object as a plot object.

          :param pm: the PrettyMIDI instance to plot
          :return: the bokeh plot layout
        """
        preset = self._preset

        # Calculates the QPM from the MIDI file, might raise exception if confused
        qpm = self._get_qpm(pm)

        # Initialize the tools, those are present on the right hand side
        plot = bokeh.plotting.figure(
            tools="reset,hover,save,wheel_zoom,pan",
            toolbar_location=preset.toolbar_location)

        # Setup the hover and the data dict for bokeh,
        # each property must match a property in the data dict
        plot.select(dict(type=bokeh.models.HoverTool)).tooltips = ({
            "program": "@program",
            "pitch": "@top",
            "velocity": "@velocity",
            "duration": "@duration",
            "start_time": "@left",
            "end_time": "@right"})
        data = dict(
            program=[],
            top=[],
            bottom=[],
            left=[],
            right=[],
            duration=[],
            velocity=[],
            color=[])

        # Puts the notes in the dict for bokeh and saves first
        # and last note time, bigger and smaller pitch
        pitch_min = None
        pitch_max = None
        first_note_start = None
        last_note_end = None
        index_instrument = 0
        for instrument in pm.instruments:
            for note in instrument.notes:
                pitch_min = min(pitch_min or self._MAX_PITCH, note.pitch)
                pitch_max = max(pitch_max or self._MIN_PITCH, note.pitch)
                color = self._get_color(index_instrument, note)
                note_start = note.start
                note_end = note.start + (note.end - note.start)
                data["program"].append(instrument.program)
                data["top"].append(note.pitch)
                if self._show_velocity:
                    data["bottom"].append(note.pitch + (note.velocity / 127))
                else:
                    data["bottom"].append(note.pitch + 1)
                data["left"].append(note_start)
                data["right"].append(note_end)
                data["duration"].append(note_end - note_start)
                data["velocity"].append(note.velocity)
                data["color"].append(color)
                first_note_start = min(first_note_start or sys.maxsize, note_start)
                last_note_end = max(last_note_end or 0, note_end)
            index_instrument = index_instrument + 1

        # Shows an empty plot even if there are no notes
        if first_note_start is None or last_note_end is None or pitch_min is None or pitch_max is None:
            pitch_min = self._MIN_PITCH
            pitch_max = pitch_min + 5
            first_note_start = 0
            last_note_end = 0

        # Gets the pitch range (min, max) from either the provided arguments
        # or min and max values from the notes
        if self._plot_pitch_range_start is not None:
            pitch_min = self._plot_pitch_range_start
        else:
            pitch_min = min(self._MAX_PITCH, pitch_min)
        if self._plot_pitch_range_stop is not None:
            pitch_max = self._plot_pitch_range_stop
        else:
            pitch_max = max(self._MIN_PITCH, pitch_max)

        pitch_range = pitch_max + 1 - pitch_min

        # Draws the rectangles on the plot from the data
        source = ColumnDataSource(data=data)
        plot.quad(left="left",
                  right="right",
                  top="top",
                  bottom="bottom",
                  line_alpha=1,
                  line_color="black",
                  color="color",
                  source=source)

        # Draws the y grid by hand, because the grid has label on the ticks, but
        # for a plot like this, the labels needs to fit in between the ticks.
        # Also useful to change the background of the grid each line
        for pitch in range(pitch_min, pitch_max + 1):
            # Draws the background box and contours, on the underlay layer, so
            # that the rectangles and over the box annotations
            fill_alpha = (0.15 if pitch % 2 == 0 else 0.00)
            box = BoxAnnotation(bottom=pitch,
                                top=pitch + 1,
                                fill_color="gray",
                                fill_alpha=fill_alpha,
                                line_color="black",
                                line_alpha=0.3,
                                line_width=1,
                                level="underlay")
            plot.add_layout(box)
            label = Label(
                x=preset.label_y_axis_offset_x,
                y=pitch + preset.label_y_axis_offset_y,
                x_units="screen",
                text=str(pitch),
                render_mode="css",
                text_font_size=preset.label_text_font_size,
                text_font_style=preset.label_text_font_style)
            plot.add_layout(label)

        # Gets the time signature from pretty midi, or 4/4 if none
        if self._midi_time_signature:
            numerator, denominator = self._midi_time_signature.split("/")
            time_signature = TimeSignature(int(numerator), int(denominator), 0)
        else:
            if pm.time_signature_changes:
                if len(pm.time_signature_changes) > 1:
                    raise Exception("Multiple time signatures are not supported")
                time_signature = pm.time_signature_changes[0]
            else:
                time_signature = TimeSignature(4, 4, 0)

        # Gets seconds per bar and seconds per beat
        if len(pm.get_beats()) >= 2:
            seconds_per_beat = pm.get_beats()[1] - pm.get_beats()[0]
        else:
            seconds_per_beat = 0.5
        if len(pm.get_downbeats()) >= 2:
            seconds_per_bar = pm.get_downbeats()[1] - pm.get_downbeats()[0]
        else:
            seconds_per_bar = 2.0

        # Defines the end time of the plot in seconds
        if self._plot_bar_range_stop is not None:
            plot_end_time = self._plot_bar_range_stop * seconds_per_bar
        else:
            # Calculates the plot start and end time, the start time can start after
            # notes or truncate notes if the plot is too long (we left truncate the
            # plot with the bounds)
            # The plot start and plot end are a multiple of seconds per bar (closest
            # smaller value for the start time, closest higher value for the end time)
            plot_end_time = int((last_note_end) / seconds_per_bar) * seconds_per_bar
            # If the last note end is exactly on a multiple of seconds per bar,
            # we don't start a new one
            is_on_bar = math.isclose(last_note_end % seconds_per_bar, seconds_per_bar)
            is_on_bar_exact = math.isclose(last_note_end % seconds_per_bar, 0.0)
            if not is_on_bar and not is_on_bar_exact:
                plot_end_time += seconds_per_bar

        # Defines the start time of the plot in seconds
        if self._plot_bar_range_start is not None:
            plot_start_time = self._plot_bar_range_start * seconds_per_bar
        else:
            start_time = int(first_note_start / seconds_per_bar) * seconds_per_bar
            plot_max_length_time = self._plot_max_length_bar * seconds_per_bar
            plot_start_time = max(plot_end_time - plot_max_length_time, start_time)

        # Draws the vertical bar grid, with a different background color
        # for each bar
        if preset.show_bar:
            bar_count = 0
            for bar_time in pm.get_downbeats():
                fill_alpha_index = bar_count % len(self._bar_fill_alphas)
                fill_alpha = self._bar_fill_alphas[fill_alpha_index]
                box = BoxAnnotation(left=bar_time,
                                    right=bar_time + seconds_per_bar,
                                    fill_color="gray",
                                    fill_alpha=fill_alpha,
                                    line_color="black",
                                    line_width=2,
                                    line_alpha=0.5,
                                    level="underlay")
                plot.add_layout(box)
                bar_count += 1

        # Draws the vertical beat grid, those are only grid lines
        if preset.show_beat:
            for beat_time in pm.get_beats():
                box = BoxAnnotation(left=beat_time,
                                    right=beat_time + seconds_per_beat,
                                    fill_color=None,
                                    line_color="black",
                                    line_width=1,
                                    line_alpha=0.4,
                                    level="underlay")
                plot.add_layout(box)

        # Configure x axis
        plot.xaxis.bounds = (plot_start_time, plot_end_time)
        plot.xaxis.axis_label = "time (SEC)"
        plot.xaxis.axis_label_text_font_size = preset.axis_label_text_font_size
        plot.xaxis.ticker = bokeh.models.SingleIntervalTicker(interval=1)
        plot.xaxis.major_tick_line_alpha = 0.9
        plot.xaxis.major_tick_line_width = 1
        plot.xaxis.major_tick_out = preset.axis_x_major_tick_out
        plot.xaxis.minor_tick_line_alpha = 0
        plot.xaxis.major_label_text_font_size = preset.label_text_font_size
        plot.xaxis.major_label_text_font_style = preset.label_text_font_style

        # Configure y axis
        plot.yaxis.bounds = (pitch_min, pitch_max + 1)
        plot.yaxis.axis_label = "pitch (MIDI)"
        plot.yaxis.axis_label_text_font_size = preset.axis_label_text_font_size
        plot.yaxis.ticker = bokeh.models.SingleIntervalTicker(interval=1)
        plot.yaxis.major_label_text_alpha = 0
        plot.yaxis.major_tick_line_alpha = 0.9
        plot.yaxis.major_tick_line_width = 1
        plot.yaxis.major_tick_out = preset.axis_y_major_tick_out
        plot.yaxis.minor_tick_line_alpha = 0
        plot.yaxis.axis_label_standoff = preset.axis_y_label_standoff
        plot.outline_line_width = 1
        plot.outline_line_alpha = 1
        plot.outline_line_color = "black"

        # The x grid is deactivated because is draw by hand (see x grid code)
        plot.xgrid.grid_line_color = None

        # The y grid is deactivated because is draw by hand (see y grid code)
        plot.ygrid.grid_line_color = None

        # Configure the plot size and range
        if self._plot_title is None:
            plot_title_text = "Visual MIDI (%s QPM, %s/%s)" % (
                str(int(qpm)), time_signature.numerator, time_signature.denominator)
        else:
            plot_title_text = self._plot_title
        
        plot.title = Title(text=plot_title_text,
                           text_font_size=preset.title_text_font_size)
        plot.plot_width = preset.plot_width
        if preset.row_height:
            plot.plot_height = pitch_range * preset.row_height
        else:
            plot.plot_height = preset.plot_height
        plot.x_range = Range1d(plot_start_time, plot_end_time)
        plot.y_range = Range1d(pitch_min, pitch_max + 1)
        plot.min_border_right = 50

        if self._live_reload and preset.stop_live_reload_button:
            callback = CustomJS(code="clearInterval(liveReloadInterval)")
            button = Button(label="stop live reload")
            button.js_on_click(callback)
            layout = column(button, plot)
        else:
            layout = column(plot)

        return layout

    def save(self, pm: PrettyMIDI, filepath: str):

def plot_growth(p,t,time_data):

    plot_var(p,t,time_data, 'X', color='black')
    p.yaxis.axis_label = "Cell concentration $[g.L^{-1}]$"
    p.y_range.start = 0.9 * time_data.loc['X'].min()
    p.y_range.end =  1.09 * time_data.loc['X'].max()

    # Setting the second y axis range name and range
    p.extra_y_ranges = {"mu": Range1d()}

    # Adding the second axis to the plot.
    p.add_layout(LinearAxis(y_range_name="mu",
                            axis_label='growth rate $[h^{-1}]$'), 'right')

    plot_var(p,t,time_data, 'mu', color='gray', y_range_name='mu')

    return p


def plot_dynamics(model, time_data):

def make_growth_plot(model,time_data):
    t = time_data.loc['t']
    y1 = time_data.loc['X']
    y2 = time_data.loc['mu']

    p = bp.figure(width=1000)
    p.title.text = 'Growth, Cell concentration over time'
    p.line(t,y1, color='black', line_width=LINE_WIDTH)

    p.y_range.start = -0.05 * y1.min()
    p.y_range.end   =  1.05 * y1.max()

    # Setting the second y axis range name and range
    p.extra_y_ranges = {"mu": Range1d(start=-0.05*y2.min(), end = 1.05*y2.max())}

    # Adding the second axis to the plot.
    p.add_layout(LinearAxis(y_range_name="mu",
                            axis_label='growth rate $[h^{-1}]$'), 'right')

    p.line(t,y2, color='grey',  line_width=LINE_WIDTH,
           line_dash = 'dashed', y_range_name='mu')

    p.xaxis.major_label_text_font_size = AXIS_FONT_SIZE
    p.yaxis.major_label_text_font_size = AXIS_FONT_SIZE
    # p.legend.label_text_font_size = LEGEND_FONT_SIZE

    return p

def get_y(the_var, time_data):

def test_bokeh(matrices):

    p = figure(plot_width=550, plot_height=550)
    xlim = (-69.551894338989271, 64.381507070922851)
    ylim = (-70.038440855407714, 70.644477995300306)
    p.axis.visible = False
    p.border_fill_color = None
    p.outline_line_color = None
    p.grid.visible = None
    p.toolbar.logo = None
    p.toolbar_location = None
    p.x_range = Range1d(start=xlim[0],end =xlim[1])
    p.y_range = Range1d(start=ylim[0],end =ylim[1])

    # p.image_url(url=['color_scatter.png'], x=0, y=0, w=None, h=None, anchor="center")
    p.image_url(url=['static/images/tsne'+LATEST_DATE+'_sml.png'], x=xlim[0]-2.0, y=ylim[1], w=ylim[1]-ylim[0]-2.7, h=(xlim[1]-xlim[0])+6.7, anchor="top_left")

    for i, matrix in enumerate(matrices):
        if len(matrix['score']) > 0:
            p.scatter(matrix['vectors'][:, 0], matrix['vectors'][:, 1],
                      fill_color='#fc6e2d', fill_alpha=0.8,
                      line_color=None, size=2)
        else:
            p.scatter(matrix['vectors'][:, 0], matrix['vectors'][:, 1],
                      fill_color='#5f4af9', fill_alpha=0.8,
                      line_color=None, size=2.5)

    # export_png(p, filename='color_scatter.png')

    # output_file("color_scatter.html", title="color_scatter.py example")
    comp = components(p)
    return comp[0]+comp[1]
    # return file_html(p, CDN)
    # save(p)



if __name__ == "__main__":

def _plot_twinx_bokeh(plot, _):
    """Hook to plot data on a secondary (twin) axis on a Holoviews Plot with Bokeh backend.

    Args:
        plot: Holoviews plot object to hook for twinx

    See Also:
        The code was copied from a comment in https://github.com/holoviz/holoviews/issues/396.
        - http://holoviews.org/user_guide/Customizing_Plots.html#plot-hooks
        - https://docs.bokeh.org/en/latest/docs/user_guide/plotting.html#twin-axes

    """
    fig: Figure = plot.state
    glyph_first: GlyphRenderer = fig.renderers[0]  # will be the original plot
    glyph_last: GlyphRenderer = fig.renderers[-1]  # will be the new plot
    right_axis_name = "twiny"
    # Create both axes if right axis does not exist
    if right_axis_name not in fig.extra_y_ranges.keys():
        # Recreate primary axis (left)
        y_first_name = glyph_first.glyph.y
        y_first_min = glyph_first.data_source.data[y_first_name].min()
        y_first_max = glyph_first.data_source.data[y_first_name].max()
        y_first_offset = (y_first_max - y_first_min) * 0.1
        fig.y_range = Range1d(
            start=y_first_min - y_first_offset,
            end=y_first_max + y_first_offset
        )
        fig.y_range.name = glyph_first.y_range_name
        # Create secondary axis (right)
        y_last_name = glyph_last.glyph.y
        y_last_min = glyph_last.data_source.data[y_last_name].min()
        y_last_max = glyph_last.data_source.data[y_last_name].max()
        y_last_offset = (y_last_max - y_last_min) * 0.1
        fig.extra_y_ranges = {right_axis_name: Range1d(
            start=y_last_min - y_last_offset,
            end=y_last_max + y_last_offset
        )}
        fig.add_layout(LinearAxis(y_range_name=right_axis_name, axis_label=glyph_last.glyph.y), "right")
    # Set right axis for the last glyph added to the figure
    glyph_last.y_range_name = right_axis_name


def get_extent_2d(shape, spacing: Optional[float] = None):

def _modify_plot(plot, figargs):
    '''Add text and modify figure attributes. This is an internal
    function which allows for figure attributes to be passed into
    interactive functions.

    Args:
        plot (bokeh.figure): The figure to modify
        figargs (dict[assorted]): A dictionary of width, height,
                title, x_axis, y_axis, x_range and y_range.

    '''
    if figargs['width'] is not None:
        plot.width = figargs['width']
    if figargs['height'] is not None:
        plot.height = figargs['height']

    if figargs['title'] is not None:
        plot.title.text = figargs['title']
    if figargs['x_axis'] is not None:
        plot.xaxis.axis_label = figargs['x_axis']
    if figargs['y_axis'] is not None:
        plot.yaxis.axis_label = figargs['y_axis']

    if figargs['x_range'] is not None:
        plot.x_range = Range1d(figargs['x_range'][0], figargs['x_range'][1])
    if figargs['y_range'] is not None:
        plot.y_range = Range1d(figargs['y_range'][0], figargs['y_range'][1])

    return plot


def show(plot, png=False, static=False, hover=True,

def dendrogram(D, figargs=None):
    '''Creates a dendrogram plot.
    This plot can show full structure of a given dendrogram.

    Args:
        D (henchman.selection.Dendrogram): An initialized dendrogram object

    Examples:
        >>> from henchman.selection import Dendrogram
        >>> from henchman.plotting import show
        >>> import henchman.plotting as hplot
        >>> D = Dendrogram(X)
        >>> plot = hplot.dendrogram(D)
        >>> show(plot)
    '''
    if figargs is None:
        return lambda figargs: dendrogram(D, figargs=figargs)
    G = nx.Graph()

    vertices_source = ColumnDataSource(
        pd.DataFrame({'index': D.columns.keys(),
                      'desc': list(D.columns.values())}))
    edges_source = ColumnDataSource(
        pd.DataFrame(D.edges[0]).rename(
            columns={1: 'end', 0: 'start'}))
    step_source = ColumnDataSource(
        pd.DataFrame({'step': [0],
                      'thresh': [D.threshlist[0]],
                      'components': [len(D.graphs[0])]}))

    G.add_nodes_from([str(x) for x in vertices_source.data['index']])
    G.add_edges_from(zip(
        [str(x) for x in edges_source.data['start']],
        [str(x) for x in edges_source.data['end']]))

    graph_renderer = from_networkx(G, nx.circular_layout,
                                   scale=1, center=(0, 0))

    graph_renderer.node_renderer.data_source = vertices_source
    graph_renderer.node_renderer.view = CDSView(source=vertices_source)
    graph_renderer.edge_renderer.data_source = edges_source
    graph_renderer.edge_renderer.view = CDSView(source=edges_source)

    plot = Plot(plot_width=400, plot_height=400,
                x_range=Range1d(-1.1, 1.1),
                y_range=Range1d(-1.1, 1.1))
    plot.title.text = "Feature Connectivity"
    graph_renderer.node_renderer.glyph = Circle(
        size=5, fill_color=Spectral4[0])
    graph_renderer.node_renderer.selection_glyph = Circle(
        size=15, fill_color=Spectral4[2])
    graph_renderer.edge_renderer.data_source = edges_source
    graph_renderer.edge_renderer.glyph = MultiLine(line_color="#CCCCCC",
                                                   line_alpha=0.6,
                                                   line_width=.5)
    graph_renderer.edge_renderer.selection_glyph = MultiLine(
        line_color=Spectral4[2],
        line_width=3)
    graph_renderer.node_renderer.hover_glyph = Circle(
        size=5,
        fill_color=Spectral4[1])
    graph_renderer.selection_policy = NodesAndLinkedEdges()
    graph_renderer.inspection_policy = NodesAndLinkedEdges()

    plot.renderers.append(graph_renderer)

    plot.add_tools(
        HoverTool(tooltips=[("feature", "@desc"),
                            ("index", "@index"), ]),
        TapTool(),
        BoxZoomTool(),
        SaveTool(),
        ResetTool())

    plot = _modify_plot(plot, figargs)

    if figargs['static']:
        return plot

    def modify_doc(doc, D, figargs):
        data_table = DataTable(source=step_source,
                               columns=[TableColumn(field='step',
                                                    title='Step'),
                                        TableColumn(field='thresh',
                                                    title='Thresh'),
                                        TableColumn(field='components',
                                                    title='Components')],
                               height=50, width=400)

        def callback(attr, old, new):
            try:
                edges = D.edges[slider.value]
                edges_source.data = ColumnDataSource(
                    pd.DataFrame(edges).rename(columns={1: 'end',
                                                        0: 'start'})).data
                step_source.data = ColumnDataSource(
                    {'step': [slider.value],
                     'thresh': [D.threshlist[slider.value]],
                     'components': [len(D.graphs[slider.value])]}).data
            except Exception as e:
                print(e)

        slider = Slider(start=0,
                        end=(len(D.edges) - 1),
                        value=0,
                        step=1,
                        title="Step")
        slider.on_change('value', callback)

        doc.add_root(column(slider, data_table, plot))
    return lambda doc: modify_doc(doc, D, figargs)


def f1(X, y, model, n_precs=1000, n_splits=1, figargs=None):

def getTrackMap(target, reference=None, mode='speed', view='lapsdelta'):
    if reference is None:
        df_, df_r = target, None
    else:
        df_, df_r = acctelemetry.lapdelta(reference, target)

    ds = ColumnDataSource(df_)
    p0 = figure(plot_height=400, plot_width=800,
                tools="crosshair,pan,reset,save,wheel_zoom")

    colors = itertools.cycle(palette)
    col0, col1 = next(colors), next(colors)

    # create the velo vs dist plot
    r0 = p0.line(x='dist_lap', y='speedkmh', source=ds, color=col0, line_width=2)

    # overwrite the (non)selection glyphs with the base line style
    # the style for the hover will be set below
    nonselected_ = Line(line_alpha=1, line_color=col0, line_width=2)
    r0.selection_glyph = nonselected_
    r0.nonselection_glyph = nonselected_

    if reference is not None:
        # create the dt vs dist plot with extra y axis, set the (non)selection glyphs
        lim = max(df_.dt.abs())
        lim += lim*.2
        p0.extra_y_ranges = {"dt": Range1d(start=-lim, end=lim)}
        p0.add_layout(LinearAxis(
            y_range_name='dt',
            axis_label='dt [s]'), 'right')
        r1 = p0.line(x='dist_lap', y='dt', source=ds, y_range_name='dt', color=col1, line_width=2)
        r1.selection_glyph = Line(line_alpha=1, line_color='red', line_width=5)
        r1.nonselection_glyph = Line(line_alpha=1, line_color=col1, line_width=2)

        # create reference velo vs dist plot
        p0.line(df_r.dist_lap, df_r.speedkmh, color=next(colors), line_width=2)

    # create an invisible renderer for velo vs dist
    # this is used to trigger the hover, thus the size is large
    c0 = p0.circle(x='dist_lap', y='speedkmh', source=ds, size=10, fill_alpha=0.0, alpha=0.0)
    c0.selection_glyph = Circle(fill_color='red', fill_alpha=1., line_color=None)
    c0.nonselection_glyph = Circle(fill_alpha=0, line_color=None)

    # create figure for track map
    p1 = figure(plot_height=400, plot_width=800, tools="crosshair,pan,reset,save,wheel_zoom")

    # create map of the track
    c1 = getLapFigure(p1, df_, ds, mode, hasref=(reference is not None))

    # add some lap tangents to guide the eye when comparing map and refmap
    if reference is not None and mode not in ['absolut', 'gainloss']:
        x0 = df_.x.values
        y0 = df_.y.values
        h = df_.heading.values
        x1 = x0 + 30*np.cos(h+np.pi/2)
        y1 = y0 + 30*np.sin(h+np.pi/2)
        p1.segment(x0=x0, y0=y0, x1=x1, y1=y1, color="#F4A582", line_width=1)

        # calculate points for the reference map drawn 'outside' of the other track map
        getLapFigure(p1, df_, ds , mode, ref=True)

    # Toooltips that show some information for each point, triggered via slider.onchange JS
    tools = ['time','dist','speedkmh']
    if reference is not None:
        tools.append('speedkmh_r')
        if mode not in ['absolut', 'gainloss', 'pedals', 'speed']:
            tools.extend([mode, '%s_r'%mode])
        elif mode in ['pedals']:
            tools.extend(['throttle', 'throttle_r', 'brake', 'brake_r'])
        tools.append('dt')
    elif mode == 'pedals':
        tools.extend(['throttle', 'brake'])
    elif mode != 'speed':
        tools.append(mode)

    hover0 = createHoverTool(tools)
    # a small hack to show only one tooltip (hover selects multiple points)
    # TODO: doesn't work anymore with recent bokeh
    # hover0.tooltips[-1] = (hover0.tooltips[-1][0], hover0.tooltips[-1][1]+"""
    #       <  style>
    #         .bk-tooltip>div:not(:first-child) {display:none;}
    #      < /style>""")
    hover0.renderers = [r0]
    hover0.mode = 'vline'
    hover0.line_policy='interp'

    # selection change via button and slider. Tooltips 'hover0' will be rendered in 'p0' using rederer 'r0'
    slider = getLapSlider(ds, p0, r0, hover0, view=view)
    btns = getLapControls(ds, slider)

    # Hovertools, that emit a selection change by modifying the slider value
    callback = CustomJS(args=dict(slider=slider), code=
    """
    let val = cb_data['index'].indices[0]
    if (val!=0 && !isNaN(val))
        slider.value = cb_data['index'].indices[0];
    """)
    p0.add_tools(HoverTool(tooltips=None, renderers=[c0],
                           callback=callback,
                           line_policy='interp', mode='vline'))
    p1.add_tools(HoverTool(tooltips=None, renderers=[c1],
                           callback=callback,
                           line_policy='interp', mode='mouse', point_policy='snap_to_data'))

    p0.add_tools(hover0)
    # p1.add_tools(hover1)

    return column(btns, slider, p0, p1, sizing_mode='scale_width')




def getTrackMapPanel(df):

    def __init__(self, mode='face-on', center=(0, 0) * u.kpc, radius=90750 * u.lyr, unit=u.kpc, coord='galactic',
                 annotation=True, rot90=0, grayscale=False, r0=8.125):
        super().__init__(grayscale=grayscale, 
                         annotation=annotation, 
                         rot90=rot90, 
                         coord=coord, 
                         mode=mode, 
                         r0=r0, 
                         center=center, 
                         radius=radius, 
                         unit=unit, 
                         figsize=None, 
                         dpi=None)
        
        # prepossessing procedure
        self._unit_english = self._unit.short_names[0]
        if self._center.unit is not None and self._radius.unit is not None:
            self._center = self._center.to(self._unit)
            self._radius = self._radius.to(self._unit)
                
        self.images_read()
        self.s = 1.0
        
        TOOLS = "pan, wheel_zoom, box_zoom, reset, save, box_select"
        
        self.bokeh_fig = figure(title="", tools=TOOLS,
                                x_range=Range1d(self._ext[0], self._ext[1], 
                                                bounds=[min(self._ext[0], self._ext[1]), max(self._ext[0], self._ext[1])]), 
                                y_range=Range1d(self._ext[2], self._ext[3], 
                                                bounds=[min(self._ext[2], self._ext[3]), max(self._ext[2], self._ext[3])]), 
                                width=1000, height=1000)
        if requests.head(self._gh_img_url, allow_redirects=True).status_code == -9999:  # connection successful
            # disabled currently because rotation and grayscale wont work
            self.bokeh_fig.image_url(url=[self._gh_img_url], 
                                     x=self._ext[0], y=self._ext[2], 
                                     w=abs(self._ext[1]-self._ext[0]), h=abs(self._ext[3]-self._ext[2]), anchor="bottom_left")
        else:
            self._img = to_bokeh_img(self._img)
            self.bokeh_fig.image_rgba(image=[self._img], x=self._ext[0], y=self._ext[2], 
                                      dw=abs(self._ext[1]-self._ext[0]), dh=abs(self._ext[3]-self._ext[2]))
        
        self.bokeh_fig.xaxis.axis_label = f'{self._coord_english} ({self._unit_english})'
        self.bokeh_fig.yaxis.axis_label = f'{self._coord_english} ({self._unit_english})'
        
    def scatter(self, x, y, *args, **kwargs):

    def __init__(self, center=(0, 0) * u.deg, radius=(180, 90) * u.deg, grayscale=False):
        super().__init__(grayscale=grayscale, 
                         projection='equirectangular', 
                         center=center, 
                         radius=radius, 
                         figsize=None, 
                         dpi=None)
        self._unit = u.degree
        self.s = 1.

        #preprocessing
        if self._center.unit is not None and self._radius.unit is not None:
            self._center = self._center.to(self._unit)
            self._radius = self._radius.to(self._unit)

        if (self._center[0] + self._radius[0]).value > 180 or (self._center[0] - self._radius[0]).value   <   -180:
            raise ValueError("The border of the width will be outside the range of -180 to 180 which is not allowed\n")
        if (self._center[1] + self._radius[1]).value > 90 or (self._center[1] - self._radius[1]).value  <  -90:
            raise ValueError("The border of the height will be outside the range of -90 to 90 which is not allowed")
        if self._radius[0]  < = 0 or self._radius[0]  < = 0:
            raise ValueError("Radius cannot be negative or 0")

        self.images_read()
        self.s = 1.0

        TOOLS = "pan, wheel_zoom, box_zoom, reset, save, box_select"
        
        self.bokeh_fig = figure(title="", tools=TOOLS,
                                x_range=Range1d(self._ext[0], self._ext[1], 
                                                bounds=[min(self._ext[0], self._ext[1]), max(self._ext[0], self._ext[1])]), 
                                y_range=Range1d(self._ext[2], self._ext[3], 
                                                bounds=[min(self._ext[2], self._ext[3]), max(self._ext[2], self._ext[3])]), 
                                width=1000, height=500)
        
        if requests.head(self._gh_img_url, allow_redirects=True).status_code == -9999:  # connection successful
            # disabled currently because rotation and grayscale wont work
            self.bokeh_fig.image_url(url=[self._gh_img_url], 
                                     x=self._ext[0], y=self._ext[2], 
                                     w=abs(self._ext[1]-self._ext[0]), h=abs(self._ext[3]-self._ext[2]), anchor="bottom_left")
        else:
            self._img = to_bokeh_img(self._img)
            self.bokeh_fig.image_rgba(image=[self._img], x=self._ext[0], y=self._ext[2], 
                                      dw=abs(self._ext[1]-self._ext[0]), dh=abs(self._ext[3]-self._ext[2]))
        
        self.bokeh_fig.xaxis.axis_label = 'Galactic Longitude (Degree)'
        self.bokeh_fig.yaxis.axis_label = 'Galactic Latitude (Degree)'
        
    def scatter(self, ra, dec, *args, **kwargs):

def get_ranges(pos):
    """
    Return appropriate range of x and y from position dict.

    Usage:
        >>> pos = nx.nx_agraph.graphviz_layout(g, prog="dot")
        >>> plot = Plot(plot_width=1000, plot_height=600, **get_ranges(pos))
    """
    all_pos = np.array(list(zip(*pos.values())))
    max_x, max_y = all_pos.max(axis=1)
    min_x, min_y = all_pos.min(axis=1)
    x_margin = max((max_x - min_x) / 10, 0.5)
    y_margin = max((max_y - min_y) / 10, 0.5)
    return {
        "x_range": Range1d(min_x - x_margin, max_x + x_margin),
        "y_range": Range1d(min_y - y_margin, max_y + y_margin),
    }


# Hacky, probably worth re-implementing
def set_edge_alpha(g):

def _plot_manual_exercise(df):
    pallete = palletes[max(palletes)] # get last

    # todo one axis for count, one for seconds? although not really gonna work for
    # maybe on separate plots?
    has_err = df['error'].notna()
    errs = df[has_err].copy()
    errs['reps'] = 1 # meh
    some_dt = df['dt'].dropna().iloc[-1]
    errs['dt'].fillna(some_dt, inplace=True)
    # not sure? some errs have reps.. errs['reps'].fillna(-5) # meh
    df  = df[~has_err]
    # TODO would be nice to reuse stuff to display errors as a table

    # FIXME handle none carefully here, otherwise they aren't displayed
    plots = []
    # todo hmm, reuse group hints somehow? not sure..

    # TODO helper groupby to check for None (to make sure they are handled)
    groups = list(df.groupby('kind'))
    # sort by the most recent
    groups = list(sorted(groups, key=lambda kind_d: max(unlocalize(kind_d[1]['dt'])), reverse=True))

    kinds = [kind for kind, _ in groups]
    # make colors stable
    colors = {kind: c for kind, c in zip(kinds, cycle(pallete))}
    colors['errors'] = 'red'

    x_range = None
    for k, edf in chain(
            [('errors', errs)],
            groups,
    ):
        color = colors[k]
        # TODO add some x jitter to declutter?
        # need to preserve the order though? I guess need to group
        p = date_figure(df, height=150, x_range=x_range)
        p.scatter(x='dt', y='reps'  , source=CDS(edf), legend_label='reps'  , color=color)

        from bokeh.models import LinearAxis, Range1d
        maxy = np.nanmax(edf['volume'] * 1.1) # TODO meh
        if not np.isnan(maxy): # I guess doesn't have volume?
            p.extra_y_ranges = {'volume': Range1d(start=0.0, end=maxy)}
            # add the second axis to the plot.
            p.add_layout(LinearAxis(y_range_name='volume'), 'right')
            p.scatter(x='dt', y='volume', source=CDS(edf), legend_label='volume', color='black', size=2, y_range_name='volume')

        p.title.text = k

        p.y_range.start = 0

        if x_range is None:
            x_range = p.x_range
        plots.append(p)
        # TODO not sure if I want sliding averages?
    return column(plots)


def _plot_strength_volume(df) -> RollingResult:

def _plot_sleep(df):
    df = _sleep_df(df)

    # TODO https://github.com/bokeh/bokeh/blob/master/examples/app/sliders.py

    # todo naming: plot vs figure vs graph?
    r = rolling(x='date', y='avg_hr', df=df, color='blue', legend_label='HR')
    [g, g7, g30] = r
    g7 .glyph.line_width = 2
    g30.glyph.line_color = 'lightblue'
    g30.glyph.line_width = 2
    p = r.figure

    p.extra_y_ranges = {'resp': Range1d(start=10, end=25)}
    # Add the second axis to the plot.
    p.add_layout(LinearAxis(y_range_name='resp'), 'right')
    rolling(
        df=df,
        x='date', y='respiratory_rate_avg',
        color='orange',
        legend_label='Respiration',
        avgs=['7D'],
        y_range_name='resp',
        context=r,
    )
    # TODO hmm, it appends error twice? not sure about it...
    add_daysoff(p)
    return r

# todo always set output_file("xx.html")? so the latest html is always dumped?


# todo what's the benefit of having date index at all?
# todo think of better naming
def plot_sleep_hrv(df):

def plot(*, results: pd.Series,
         df: pd.DataFrame,
         indicators: List[_Indicator],
         filename='', plot_width=None,
         plot_equity=True, plot_return=False, plot_pl=True,
         plot_volume=True, plot_drawdown=False,
         smooth_equity=False, relative_equity=True,
         superimpose=True, resample=True,
         reverse_indicators=True,
         show_legend=True, open_browser=True):
    """
    Like much of GUI code everywhere, this is a mess.
    """
    # We need to reset global Bokeh state, otherwise subsequent runs of
    # plot() contain some previous run's cruft data (was noticed when
    # TestPlot.test_file_size() test was failing).
    if not filename and not IS_JUPYTER_NOTEBOOK:
        filename = _windos_safe_filename(str(results._strategy))
    _bokeh_reset(filename)

    COLORS = [BEAR_COLOR, BULL_COLOR]
    BAR_WIDTH = .8

    assert df.index.equals(results['_equity_curve'].index)
    equity_data = results['_equity_curve'].copy(deep=False)
    trades = results['_trades']

    plot_volume = plot_volume and not df.Volume.isnull().all()
    plot_equity = plot_equity and not trades.empty
    plot_return = plot_return and not trades.empty
    plot_pl = plot_pl and not trades.empty
    is_datetime_index = isinstance(df.index, pd.DatetimeIndex)

    from .lib import OHLCV_AGG
    # ohlc df may contain many columns. We're only interested in, and pass on to Bokeh, these
    df = df[list(OHLCV_AGG.keys())].copy(deep=False)

    # Limit data to max_candles
    if is_datetime_index:
        df, indicators, equity_data, trades = _maybe_resample_data(
            resample, df, indicators, equity_data, trades)

    df.index.name = None  # Provides source name @index
    df['datetime'] = df.index  # Save original, maybe datetime index
    df = df.reset_index(drop=True)
    equity_data = equity_data.reset_index(drop=True)
    index = df.index

    new_bokeh_figure = partial(
        _figure,
        x_axis_type='linear',
        width=plot_width,
        height=400,
        tools="xpan,xwheel_zoom,box_zoom,undo,redo,reset,save",
        active_drag='xpan',
        active_scroll='xwheel_zoom')

    pad = (index[-1] - index[0]) / 20

    fig_ohlc = new_bokeh_figure(
        x_range=Range1d(index[0], index[-1],
                        min_interval=10,
                        bounds=(index[0] - pad,
                                index[-1] + pad)) if index.size > 1 else None)
    figs_above_ohlc, figs_below_ohlc = [], []

    source = ColumnDataSource(df)
    source.add((df.Close >= df.Open).values.astype(np.uint8).astype(str), 'inc')

    trade_source = ColumnDataSource(dict(
        index=trades['ExitBar'],
        datetime=trades['ExitTime'],
        exit_price=trades['ExitPrice'],
        size=trades['Size'],
        returns_positive=(trades['ReturnPct'] > 0).astype(int).astype(str),
    ))

    inc_cmap = factor_cmap('inc', COLORS, ['0', '1'])
    cmap = factor_cmap('returns_positive', COLORS, ['0', '1'])
    colors_darker = [lightness(BEAR_COLOR, .35),
                     lightness(BULL_COLOR, .35)]
    trades_cmap = factor_cmap('returns_positive', colors_darker, ['0', '1'])

    if is_datetime_index:
        fig_ohlc.xaxis.formatter = CustomJSTickFormatter(
            args=dict(axis=fig_ohlc.xaxis[0],
                      formatter=DatetimeTickFormatter(days=['%d %b', '%a %d'],
                                                      months=['%m/%Y', "%b'%y"]),
                      source=source),
            code='''
this.labels = this.labels || formatter.doFormat(ticks
                                                .map(i => source.data.datetime[i])
                                                .filter(t => t !== undefined));
return this.labels[index] || "";
        ''')

    NBSP = '\N{NBSP}' * 4
    ohlc_extreme_values = df[['High', 'Low']].copy(deep=False)
    ohlc_tooltips = [
        ('x, y', NBSP.join(('$index',
                            '$y{0,0.0[0000]}'))),
        ('OHLC', NBSP.join(('@Open{0,0.0[0000]}',
                            '@High{0,0.0[0000]}',
                            '@Low{0,0.0[0000]}',
                            '@Close{0,0.0[0000]}'))),
        ('Volume', '@Volume{0,0}')]

    def new_indicator_figure(**kwargs):
        kwargs.setdefault('height', 90)
        fig = new_bokeh_figure(x_range=fig_ohlc.x_range,
                               active_scroll='xwheel_zoom',
                               active_drag='xpan',
                               **kwargs)
        fig.xaxis.visible = False
        fig.yaxis.minor_tick_line_color = None
        return fig

    def set_tooltips(fig, tooltips=(), vline=True, renderers=()):
        tooltips = list(tooltips)
        renderers = list(renderers)

        if is_datetime_index:
            formatters = {'@datetime': 'datetime'}
            tooltips = [("Date", "@datetime{%c}")] + tooltips
        else:
            formatters = {}
            tooltips = [("#", "@index")] + tooltips
        fig.add_tools(HoverTool(
            point_policy='follow_mouse',
            renderers=renderers, formatters=formatters,
            tooltips=tooltips, mode='vline' if vline else 'mouse'))

    def _plot_equity_section(is_return=False):
        """Equity section"""
        # Max DD Dur. line
        equity = equity_data['Equity'].copy()
        dd_end = equity_data['DrawdownDuration'].idxmax()
        if np.isnan(dd_end):
            dd_start = dd_end = equity.index[0]
        else:
            dd_start = equity[:dd_end].idxmax()
            # If DD not extending into the future, get exact point of intersection with equity
            if dd_end != equity.index[-1]:
                dd_end = np.interp(equity[dd_start],
                                   (equity[dd_end - 1], equity[dd_end]),
                                   (dd_end - 1, dd_end))

        if smooth_equity:
            interest_points = pd.Index([
                # Beginning and end
                equity.index[0], equity.index[-1],
                # Peak equity and peak DD
                equity.idxmax(), equity_data['DrawdownPct'].idxmax(),
                # Include max dd end points. Otherwise the MaxDD line looks amiss.
                dd_start, int(dd_end), min(int(dd_end + 1), equity.size - 1),
            ])
            select = pd.Index(trades['ExitBar']).union(interest_points)
            select = select.unique().dropna()
            equity = equity.iloc[select].reindex(equity.index)
            equity.interpolate(inplace=True)

        assert equity.index.equals(equity_data.index)

        if relative_equity:
            equity /= equity.iloc[0]
        if is_return:
            equity -= equity.iloc[0]

        yaxis_label = 'Return' if is_return else 'Equity'
        source_key = 'eq_return' if is_return else 'equity'
        source.add(equity, source_key)
        fig = new_indicator_figure(
            y_axis_label=yaxis_label,
            **({} if plot_drawdown else dict(height=110)))

        # High-watermark drawdown dents
        fig.patch('index', 'equity_dd',
                  source=ColumnDataSource(dict(
                      index=np.r_[index, index[::-1]],
                      equity_dd=np.r_[equity, equity.cummax()[::-1]]
                  )),
                  fill_color='#ffffea', line_color='#ffcb66')

        # Equity line
        r = fig.line('index', source_key, source=source, line_width=1.5, line_alpha=1)
        if relative_equity:
            tooltip_format = f'@{source_key}{{+0,0.[000]%}}'
            tick_format = '0,0.[00]%'
            legend_format = '{:,.0f}%'
        else:
            tooltip_format = f'@{source_key}{{$ 0,0}}'
            tick_format = '$ 0.0 a'
            legend_format = '${:,.0f}'
        set_tooltips(fig, [(yaxis_label, tooltip_format)], renderers=[r])
        fig.yaxis.formatter = NumeralTickFormatter(format=tick_format)

        # Peaks
        argmax = equity.idxmax()
        fig.scatter(argmax, equity[argmax],
                    legend_label='Peak ({})'.format(
                        legend_format.format(equity[argmax] * (100 if relative_equity else 1))),
                    color='cyan', size=8)
        fig.scatter(index[-1], equity.values[-1],
                    legend_label='Final ({})'.format(
                        legend_format.format(equity.iloc[-1] * (100 if relative_equity else 1))),
                    color='blue', size=8)

        if not plot_drawdown:
            drawdown = equity_data['DrawdownPct']
            argmax = drawdown.idxmax()
            fig.scatter(argmax, equity[argmax],
                        legend_label='Max Drawdown (-{:.1f}%)'.format(100 * drawdown[argmax]),
                        color='red', size=8)
        dd_timedelta_label = df['datetime'].iloc[int(round(dd_end))] - df['datetime'].iloc[dd_start]
        fig.line([dd_start, dd_end], equity.iloc[dd_start],
                 line_color='red', line_width=2,
                 legend_label=f'Max Dd Dur. ({dd_timedelta_label})'
                 .replace(' 00:00:00', '')
                 .replace('(0 days ', '('))

        figs_above_ohlc.append(fig)

    def _plot_drawdown_section():
        """Drawdown section"""
        fig = new_indicator_figure(y_axis_label="Drawdown")
        drawdown = equity_data['DrawdownPct']
        argmax = drawdown.idxmax()
        source.add(drawdown, 'drawdown')
        r = fig.line('index', 'drawdown', source=source, line_width=1.3)
        fig.scatter(argmax, drawdown[argmax],
                    legend_label='Peak (-{:.1f}%)'.format(100 * drawdown[argmax]),
                    color='red', size=8)
        set_tooltips(fig, [('Drawdown', '@drawdown{-0.[0]%}')], renderers=[r])
        fig.yaxis.formatter = NumeralTickFormatter(format="-0.[0]%")
        return fig

    def _plot_pl_section():
        """Profit/Loss markers section"""
        fig = new_indicator_figure(y_axis_label="Profit / Loss")
        fig.add_layout(Span(location=0, dimension='width', line_color='#666666',
                            line_dash='dashed', line_width=1))
        returns_long = np.where(trades['Size'] > 0, trades['ReturnPct'], np.nan)
        returns_short = np.where(trades['Size']   <   0, trades['ReturnPct'], np.nan)
        size = trades['Size'].abs()
        size = np.interp(size, (size.min(), size.max()), (8, 20))
        trade_source.add(returns_long, 'returns_long')
        trade_source.add(returns_short, 'returns_short')
        trade_source.add(size, 'marker_size')
        if 'count' in trades:
            trade_source.add(trades['count'], 'count')
        r1 = fig.scatter('index', 'returns_long', source=trade_source, fill_color=cmap,
                         marker='triangle', line_color='black', size='marker_size')
        r2 = fig.scatter('index', 'returns_short', source=trade_source, fill_color=cmap,
                         marker='inverted_triangle', line_color='black', size='marker_size')
        tooltips = [("Size", "@size{0,0}")]
        if 'count' in trades:
            tooltips.append(("Count", "@count{0,0}"))
        set_tooltips(fig, tooltips + [("P/L", "@returns_long{+0.[000]%}")],
                     vline=False, renderers=[r1])
        set_tooltips(fig, tooltips + [("P/L", "@returns_short{+0.[000]%}")],
                     vline=False, renderers=[r2])
        fig.yaxis.formatter = NumeralTickFormatter(format="0.[00]%")
        return fig

    def _plot_volume_section():
        """Volume section"""
        fig = new_indicator_figure(y_axis_label="Volume")
        fig.xaxis.formatter = fig_ohlc.xaxis[0].formatter
        fig.xaxis.visible = True
        fig_ohlc.xaxis.visible = False  # Show only Volume's xaxis
        r = fig.vbar('index', BAR_WIDTH, 'Volume', source=source, color=inc_cmap)
        set_tooltips(fig, [('Volume', '@Volume{0.00 a}')], renderers=[r])
        fig.yaxis.formatter = NumeralTickFormatter(format="0 a")
        return fig

    def _plot_superimposed_ohlc():
        """Superimposed, downsampled vbars"""
        time_resolution = pd.DatetimeIndex(df['datetime']).resolution
        resample_rule = (superimpose if isinstance(superimpose, str) else
                         dict(day='M',
                              hour='D',
                              minute='H',
                              second='T',
                              millisecond='S').get(time_resolution))
        if not resample_rule:
            warnings.warn(
                f"'Can't superimpose OHLC data with rule '{resample_rule}'"
                f"(index datetime resolution: '{time_resolution}'). Skipping.",
                stacklevel=4)
            return

        df2 = (df.assign(_width=1).set_index('datetime')
               .resample(resample_rule, label='left')
               .agg(dict(OHLCV_AGG, _width='count')))

        # Check if resampling was downsampling; error on upsampling
        orig_freq = _data_period(df['datetime'])
        resample_freq = _data_period(df2.index)
        if resample_freq  <  orig_freq:
            raise ValueError('Invalid value for `superimpose`: Upsampling not supported.')
        if resample_freq == orig_freq:
            warnings.warn('Superimposed OHLC plot matches the original plot. Skipping.',
                          stacklevel=4)
            return

        df2.index = df2['_width'].cumsum().shift(1).fillna(0)
        df2.index += df2['_width'] / 2 - .5
        df2['_width'] -= .1  # Candles don't touch

        df2['inc'] = (df2.Close >= df2.Open).astype(int).astype(str)
        df2.index.name = None
        source2 = ColumnDataSource(df2)
        fig_ohlc.segment('index', 'High', 'index', 'Low', source=source2, color='#bbbbbb')
        colors_lighter = [lightness(BEAR_COLOR, .92),
                          lightness(BULL_COLOR, .92)]
        fig_ohlc.vbar('index', '_width', 'Open', 'Close', source=source2, line_color=None,
                      fill_color=factor_cmap('inc', colors_lighter, ['0', '1']))

    def _plot_ohlc():
        """Main OHLC bars"""
        fig_ohlc.segment('index', 'High', 'index', 'Low', source=source, color="black")
        r = fig_ohlc.vbar('index', BAR_WIDTH, 'Open', 'Close', source=source,
                          line_color="black", fill_color=inc_cmap)
        return r

    def _plot_ohlc_trades():
        """Trade entry / exit markers on OHLC plot"""
        trade_source.add(trades[['EntryBar', 'ExitBar']].values.tolist(), 'position_lines_xs')
        trade_source.add(trades[['EntryPrice', 'ExitPrice']].values.tolist(), 'position_lines_ys')
        fig_ohlc.multi_line(xs='position_lines_xs', ys='position_lines_ys',
                            source=trade_source, line_color=trades_cmap,
                            legend_label=f'Trades ({len(trades)})',
                            line_width=8, line_alpha=1, line_dash='dotted')

    def _plot_indicators():
        """Strategy indicators"""

        def _too_many_dims(value):
            assert value.ndim >= 2
            if value.ndim > 2:
                warnings.warn(f"Can't plot indicators with >2D ('{value.name}')",
                              stacklevel=5)
                return True
            return False

        class LegendStr(str):
            # The legend string is such a string that only matches
            # itself if it's the exact same object. This ensures
            # legend items are listed separately even when they have the
            # same string contents. Otherwise, Bokeh would always consider
            # equal strings as one and the same legend item.
            def __eq__(self, other):
                return self is other

        ohlc_colors = colorgen()
        indicator_figs = []

        for i, value in enumerate(indicators):
            value = np.atleast_2d(value)

            # Use .get()! A user might have assigned a Strategy.data-evolved
            # _Array without Strategy.I()
            if not value._opts.get('plot') or _too_many_dims(value):
                continue

            is_overlay = value._opts['overlay']
            is_scatter = value._opts['scatter']
            if is_overlay:
                fig = fig_ohlc
            else:
                fig = new_indicator_figure()
                indicator_figs.append(fig)
            tooltips = []
            colors = value._opts['color']
            colors = colors and cycle(_as_list(colors)) or (
                cycle([next(ohlc_colors)]) if is_overlay else colorgen())
            legend_label = LegendStr(value.name)
            for j, arr in enumerate(value, 1):
                color = next(colors)
                source_name = f'{legend_label}_{i}_{j}'
                if arr.dtype == bool:
                    arr = arr.astype(int)
                source.add(arr, source_name)
                tooltips.append(f'@{{{source_name}}}{{0,0.0[0000]}}')
                if is_overlay:
                    ohlc_extreme_values[source_name] = arr
                    if is_scatter:
                        fig.scatter(
                            'index', source_name, source=source,
                            legend_label=legend_label, color=color,
                            line_color='black', fill_alpha=.8,
                            marker='circle', radius=BAR_WIDTH / 2 * 1.5)
                    else:
                        fig.line(
                            'index', source_name, source=source,
                            legend_label=legend_label, line_color=color,
                            line_width=1.3)
                else:
                    if is_scatter:
                        r = fig.scatter(
                            'index', source_name, source=source,
                            legend_label=LegendStr(legend_label), color=color,
                            marker='circle', radius=BAR_WIDTH / 2 * .9)
                    else:
                        r = fig.line(
                            'index', source_name, source=source,
                            legend_label=LegendStr(legend_label), line_color=color,
                            line_width=1.3)
                    # Add dashed centerline just because
                    mean = float(pd.Series(arr).mean())
                    if not np.isnan(mean) and (abs(mean)  <  .1 or
                                               round(abs(mean), 1) == .5 or
                                               round(abs(mean), -1) in (50, 100, 200)):
                        fig.add_layout(Span(location=float(mean), dimension='width',
                                            line_color='#666666', line_dash='dashed',
                                            line_width=.5))
            if is_overlay:
                ohlc_tooltips.append((legend_label, NBSP.join(tooltips)))
            else:
                set_tooltips(fig, [(legend_label, NBSP.join(tooltips))], vline=True, renderers=[r])
                # If the sole indicator line on this figure,
                # have the legend only contain text without the glyph
                if len(value) == 1:
                    fig.legend.glyph_width = 0
        return indicator_figs

    # Construct figure ...

    if plot_equity:
        _plot_equity_section()

    if plot_return:
        _plot_equity_section(is_return=True)

    if plot_drawdown:
        figs_above_ohlc.append(_plot_drawdown_section())

    if plot_pl:
        figs_above_ohlc.append(_plot_pl_section())

    if plot_volume:
        fig_volume = _plot_volume_section()
        figs_below_ohlc.append(fig_volume)

    if superimpose and is_datetime_index:
        _plot_superimposed_ohlc()

    ohlc_bars = _plot_ohlc()
    _plot_ohlc_trades()
    indicator_figs = _plot_indicators()
    if reverse_indicators:
        indicator_figs = indicator_figs[::-1]
    figs_below_ohlc.extend(indicator_figs)

    set_tooltips(fig_ohlc, ohlc_tooltips, vline=True, renderers=[ohlc_bars])

    source.add(ohlc_extreme_values.min(1), 'ohlc_low')
    source.add(ohlc_extreme_values.max(1), 'ohlc_high')

    custom_js_args = dict(ohlc_range=fig_ohlc.y_range,
                          source=source)
    if plot_volume:
        custom_js_args.update(volume_range=fig_volume.y_range)

    fig_ohlc.x_range.js_on_change('end', CustomJS(args=custom_js_args,
                                                  code=_AUTOSCALE_JS_CALLBACK))

    plots = figs_above_ohlc + [fig_ohlc] + figs_below_ohlc
    linked_crosshair = CrosshairTool(dimensions='both')

    for f in plots:
        if f.legend:
            f.legend.visible = show_legend
            f.legend.location = 'top_left'
            f.legend.border_line_width = 1
            f.legend.border_line_color = '#333333'
            f.legend.padding = 5
            f.legend.spacing = 0
            f.legend.margin = 0
            f.legend.label_text_font_size = '8pt'
            f.legend.click_policy = "hide"
        f.min_border_left = 0
        f.min_border_top = 3
        f.min_border_bottom = 6
        f.min_border_right = 10
        f.outline_line_color = '#666666'

        f.add_tools(linked_crosshair)
        wheelzoom_tool = next(wz for wz in f.tools if isinstance(wz, WheelZoomTool))
        wheelzoom_tool.maintain_focus = False

    kwargs = {}
    if plot_width is None:
        kwargs['sizing_mode'] = 'stretch_width'

    fig = gridplot(
        plots,
        ncols=1,
        toolbar_location='right',
        toolbar_options=dict(logo=None),
        merge_tools=True,
        **kwargs
    )
    show(fig, browser=None if open_browser else 'none')
    return fig


def plot_heatmaps(heatmap: pd.Series, agg: Union[Callable, str], ncols: int,

    def plot_tabs(self, show=False, **kwargs):
        """Plot the LDCs in a tabbed figure

        Parameters
        ----------
        fig: matplotlib.pyplot.figure, bokeh.plotting.figure (optional)
            An existing figure to plot on
        show: bool
            Show the figure
        """
        # Change names to reflect ld profile
        old_names = self.results['name']
        for n, row in enumerate(self.results):
            self.results[n]['name'] = row['profile']

        # Draw a figure for each wavelength bin
        tabs = []
        for wav in np.unique(self.results['wave_eff']):

            # Plot it
            TOOLS = 'box_zoom, box_select, crosshair, reset, hover'
            fig = bkp.figure(tools=TOOLS, x_range=Range1d(0, 1), y_range=Range1d(0, 1),
                        plot_width=800, plot_height=400)
            self.plot(wave_eff=wav, fig=fig)

            # Plot formatting
            fig.legend.location = 'bottom_right'
            fig.xaxis.axis_label = 'mu'
            fig.yaxis.axis_label = 'Intensity'

            tabs.append(Panel(child=fig, title=str(wav)))

        # Make the final tabbed figure
        final = Tabs(tabs=tabs)

        # Put the names back
        self.results['name'] = old_names

        if show:
            bkp.show(final)
        else:
            return final

    def plot(self, fig=None, show=False, **kwargs):

    def plot(self, fig=None, show=False, **kwargs):
        """Plot the LDCs

        Parameters
        ----------
        fig: matplotlib.pyplot.figure, bokeh.plotting.figure (optional)
            An existing figure to plot on
        show: bool
            Show the figure
        """
        # Separate plotting kwargs from parameter kwargs
        pwargs = {i: j for i, j in kwargs.items() if i in self.results.columns}
        kwargs = {i: j for i, j in kwargs.items() if i not in pwargs.keys()}

        # Filter the table by given kwargs
        table = utils.filter_table(self.results, **pwargs)

        for row in table:

            # Set color and label for plot
            color = row['color']
            label = row['name']

            # Generate smooth curve
            ldfunc = row['ldfunc']
            mu_vals = np.linspace(0, 1, 1000)
            ld_vals = ldfunc(mu_vals, *row['coeffs'])

            # Generate smooth errors
            dn_err, up_err = self.bootstrap_errors(mu_vals, ldfunc,
                                                   row['coeffs'],
                                                   row['errors'])

            # Matplotlib fig by default
            if fig is None:
                fig = bkp.figure()

            # Add fits to matplotlib
            if isinstance(fig, matplotlib.figure.Figure):

                # Make axes
                ax = fig.add_subplot(111)

                # Plot the fitted points
                ax.errorbar(row['raw_mu'], row['raw_ld'], c='k',
                            ls='None', marker='o', markeredgecolor='k',
                            markerfacecolor='None')

                # Plot the mu cutoff
                ax.axvline(row['mu_min'], color='0.5', ls=':')

                # Draw the curve and error
                ax.plot(mu_vals, ld_vals, color=color, label=label, **kwargs)
                ax.fill_between(mu_vals, dn_err, up_err, color=color,
                                alpha=0.1)
                ax.set_ylim(0, 1)
                ax.set_xlim(0, 1)

            # Or to bokeh!
            else:

                # Set the plot elements
                fig.x_range = Range1d(0, 1)
                fig.y_range = Range1d(0, 1)
                fig.xaxis.axis_label = 'mu'
                fig.yaxis.axis_label = 'Normalized Intensity'
                fig.legend.location = "bottom_right"

                # Plot the fitted points
                fig.circle(row['raw_mu'], row['raw_ld'], fill_color='black')

                # Plot the mu cutoff
                fig.line([row['mu_min']]*2, [0, 1], legend='cutoff',
                         line_color='#6b6ecf', line_dash='dotted')

                # Draw the curve and error
                fig.line(mu_vals, ld_vals, line_color=color, legend=label,
                         **kwargs)
                vals = np.append(mu_vals, mu_vals[::-1])
                evals = np.append(dn_err, up_err[::-1])
                fig.patch(vals, evals, color=color, fill_alpha=0.2,
                          line_alpha=0)

        if show:
            if isinstance(fig, matplotlib.figure.Figure):
                plt.xlabel(r'$\mu$')
                plt.ylabel(r'$I(\mu)/I(\mu = 1)$')
                plt.legend(loc=0, frameon=False)
                plt.show()
            else:
                bkp.show(fig)

        else:
            return fig

def compare_plot():

    stats = PreText(text='', width=500)
    corr = PreText(text='', width=500)
    def _ma(series,n):
        return series.rolling(window=n).mean()
    
    # connect to Cassandra database
    def make_dataset(start='2014-01-01'):
        df=pd.DataFrame()
        for comp in symbol_list:
            database=CassandraStorage(comp)
            database.session.row_factory = pandas_factory
            database.session.default_fetch_size = None
            
            query="SELECT * FROM {} WHERE time>'{}' ALLOW FILTERING;".format(database.symbol+'_historical',start)
            rslt = database.session.execute(query, timeout=None)
            df_comp = rslt._current_rows
            df_comp['ma5']=_ma(df_comp.adjusted_close,5)
            df_comp['ma10']=_ma(df_comp.adjusted_close,10)
            df_comp['ma30']=_ma(df_comp.adjusted_close,30)
            df=df.append(df_comp)
        return df
    
    
    df_all=make_dataset(start='2014-01-01')
    df_init=df_all[df_all.symbol==symbol_list[0]]
    source=ColumnDataSource(data= df_init.to_dict('list'))
    
    # hover setting
    TOOLTIPS = [
            ("time", "@time{%F}"),
            ("adjusted close", "$@adjusted_close"),
             ("close", "$@close"),
            ("open", "$@open"),
            ("high", "$@high"),
            ("low", "$@low"),
            ("volume","@volume")]
    formatters={
        'time'      : 'datetime'}
    hover = HoverTool(tooltips=TOOLTIPS,formatters=formatters,mode='vline')
    
    # create plot
    p = figure(title='{} (Click on legend entries to hide the corresponding lines)'.format(symbol_list[0]),
                plot_height=300,  
                tools="crosshair,save,undo,xpan,xwheel_zoom,ybox_zoom,reset", 
                active_scroll='xwheel_zoom',
                x_axis_type="datetime", 
                y_axis_location="left")
    
    p.add_tools(hover)
    palte=all_palettes['Set2'][8]
    p.line('time', 'adjusted_close', alpha=1.0, line_width=2, color=palte[3], legend='Adjusted Close',source=source)
    p.line('time', 'ma5', line_width=1, color=palte[0], alpha=0.8, muted_color='navy', muted_alpha=0.1, legend='MA5',source=source)
    p.line('time', 'ma10', line_width=1, color=palte[1], alpha=0.8, muted_color='navy', muted_alpha=0.1, legend='MA10',source=source)
    p.line('time', 'ma30', line_width=1, color=palte[2], alpha=0.8, muted_color='navy', muted_alpha=0.1, legend='MA30',source=source)
    p.y_range = Range1d(min(source.data['adjusted_close'])*0.3, max(source.data['adjusted_close'])*1.05)
    p.extra_y_ranges = {"volumes": Range1d(start=min(source.data['volume'])/2, 
                                           end=max(source.data['volume'])*2)}
    p.add_layout(LinearAxis(y_range_name="volumes"), 'right')
    p.vbar('time', width=3,top='volume', color=choice(all_palettes['Set2'][8]),alpha=0.5, y_range_name="volumes",source=source)
    p.legend.location = "top_left"
    p.legend.click_policy="hide"
        
    def callback(attr,old,new):
        symbol1, symbol2=compare_select1.value, compare_select2.value
        df_init1=df_all[df_all.symbol==symbol1]
        df_init2=df_all[df_all.symbol==symbol2]

        source.data.update( df_init1.to_dict('list'))
        p.title.text =symbol1+' (Click on legend entries to hide the corresponding lines)'
        p.y_range.start=min(source.data['adjusted_close'])*0.3
        p.y_range.end=max(source.data['adjusted_close'])*1.05
        p.extra_y_ranges['volumes'].start=min(source.data['volume'])/2.
        p.extra_y_ranges['volumes'].end=max(source.data['volume'])*2.

        source2.data.update( df_init2.to_dict('list'))
        p2.title.text =symbol2+' (Click on legend entries to hide the corresponding lines)'
        p2.y_range.start=min(source2.data['adjusted_close'])*0.3
        p2.y_range.end=max(source2.data['adjusted_close'])*1.05
        p2.extra_y_ranges['volumes'].start=min(source2.data['volume'])/2.
        p2.extra_y_ranges['volumes'].end=max(source2.data['volume'])*2.

        update_stat(symbol1,symbol2)
    
        
    
    def update_stat(symbol1,symbol2):
        des=pd.DataFrame()
        des[symbol1]=df_all[df_all.symbol==symbol1]['adjusted_close']
        des[symbol2]=df_all[df_all.symbol==symbol2]['adjusted_close']
        des[symbol1+'_return']=des[symbol1].pct_change()
        des[symbol2+'_return']=des[symbol2].pct_change()
        stats.text = "Statistics \n"+str(des.describe())
        
        correlation=des[[symbol1+'_return',symbol2+'_return']].corr()
        corr.text="Correlation \n"+str(correlation)
    
    update_stat(symbol_list[0],symbol_list[1])
    compare_select1=Select(value=symbol_list[0],options=symbol_list)
    compare_select1.on_change('value', callback)
    
    
    
    
    """
    Tab2-plot2: 
        
    """
    
    df_init2=df_all[df_all.symbol==symbol_list[1]]
    source2=ColumnDataSource(data= df_init2.to_dict('list'))
    
    # create plot
    # hover setting
    p2 = figure(title='{}  (Click on legend entries to hide the corresponding lines)'.format(symbol_list[1]),plot_height=300,  
                tools="crosshair,save,undo,xpan,xwheel_zoom,ybox_zoom,reset", 
                active_scroll='xwheel_zoom',
                x_axis_type="datetime", 
                y_axis_location="left")
    
    p2.add_tools(hover)
    p2.line('time', 'adjusted_close', alpha=1.0, line_width=2, color=palte[4], legend='Adjusted Close',source=source2)
    p2.line('time', 'ma5', line_width=1, color=palte[5], alpha=0.8, muted_color='navy', muted_alpha=0.1, legend='MA5',source=source2)
    p2.line('time', 'ma10', line_width=1, color=palte[6], alpha=0.8, muted_color='navy', muted_alpha=0.1, legend='MA10',source=source2)
    p2.line('time', 'ma30', line_width=1, color=palte[7], alpha=0.8, muted_color='navy', muted_alpha=0.1, legend='MA30',source=source2)

    p2.y_range = Range1d(min(source2.data['adjusted_close'])*0.3, max(source2.data['adjusted_close'])*1.05)

    
    p2.extra_y_ranges = {"volumes": Range1d(start=min(source2.data['volume'])/2, 
                                           end=max(source2.data['volume'])*2)}
    p2.add_layout(LinearAxis(y_range_name="volumes"), 'right')
    p2.vbar('time', width=3,top='volume', color=choice(all_palettes['Set2'][8]),alpha=0.5, y_range_name="volumes",source=source2)


    p2.legend.location = "top_left"
    p2.legend.click_policy="hide"
    
    compare_select2=Select(value=symbol_list[1],options=symbol_list)
    compare_select2.on_change('value', callback)
    
    widget=column(compare_select1,compare_select2)
    
    return p,p2,widget,stats,corr

def candlestick():
    if '^GSPC' in symbol_list:
        symbol_list.remove('^GSPC')
    stock_select=Select(value=symbol_list[0],options=symbol_list)
    summaryText = Div(text="",width=400)
    financialText=Div(text="",width=180)
    
    def update_summary(symbol):
        company,buzzsummary,officerString,institution_df=read_company(symbol)
        summaryText.text ="""  <  b> < p style="color:blue;">Overview:  < /p> < /b>
                              < b>Company: < /b> {} < br>
                              < b>Address: < /b> {}  < br>
                              < b>City: < /b> {}  < br>
                              < b>State: < /b> {}  < br>
                              < b>Website: < /b>  < a href="{}">{} < /a>  < br>
                              < b>Industry: < /b> {}  < br>
                              < b>Sector: < /b> {}  < br>
                              < b>Company Officers: < /b> {}  < br>                             
                              < b>Summary: < /b> {}  < br>""".format(company['price']['longName'],
                                                     company['summaryProfile']['address1'],
                                                     company['summaryProfile']['city'],
                                                     company['summaryProfile']['state'],
                                                     company['summaryProfile']['website'],
                                                     company['summaryProfile']['website'],
                                                     company['summaryProfile']['industry'],
                                                     company['summaryProfile']['sector'],
                                                     officerString,
                                                     buzzsummary)
        financialText.text=""" < b> < p style="color:blue;">Financial:  < /p> < /b>
                               < b>Recommendation: {} < /b>  < br>
                               < b>Enterprise Value: < /b> {}  < br>
                               < b>Profit Margins: < /b> {}  < br>
                               < b>Beta: < /b> {}  < br>
                               < b>EBITDA: < /b> {}  < br>
                               < b>Total Debt: < /b> {}  < br>
                               < b>Total Revenue: < /b> {} < br>
                               < b>DebtToEquity: < /b> {} < br>
                               < b>Revenue Growth: < /b> {}  < br>
                               < b>Current Ratio: < /b> {}  < br>
                               < b>ROE: < /b> {}  < br>
                               < b>ROA: < /b> {}  < br>
                               < b>Gross Profits: < /b> {}  < br>
                               < b>Quick Ratio: < /b> {}  < br>
                               < b>Free Cashflow: < /b> {}  < br>
                              """.format(company['financialData']['recommendationKey'].upper(),
                                         company['defaultKeyStatistics']['enterpriseValue']['fmt'],
                                         company['defaultKeyStatistics']['profitMargins']['fmt'],
                                         company['defaultKeyStatistics']['beta']['fmt'],
                                         company['financialData']['ebitda']['fmt'],
                                         company['financialData']['totalDebt']['fmt'],
                                         company['financialData']['totalRevenue']['fmt'],
                                         company['financialData']['debtToEquity']['fmt'],
                                         company['financialData']['revenueGrowth']['fmt'],
                                         company['financialData']['currentRatio']['fmt'],
                                         company['financialData']['returnOnAssets']['fmt'],
                                         company['financialData']['returnOnEquity']['fmt'],
                                         company['financialData']['grossProfits']['fmt'],
                                         company['financialData']['quickRatio']['fmt'],
                                         company['financialData']['freeCashflow']['fmt'])

    update_summary(stock_select.value)
    # connect to Cassandra database
    database=CassandraStorage(symbol_list[0])
    database.session.row_factory = pandas_factory
    database.session.default_fetch_size = None
    
    query="SELECT * FROM {} WHERE time>'2015-01-01'  ALLOW FILTERING;".format('{}_historical'.format(symbol_list[0]))
    rslt = database.session.execute(query, timeout=None)
    df = rslt._current_rows
    
    # create color list
    color=df.close>df.open
    color=color.replace(True,'green')
    color=color.replace(False,'red')

    # set data source
    source = ColumnDataSource(data=dict(close=list(df.close.values),
                                        adjusted_close=list(df.adjusted_close.values),
                                        open=list(df.open.values),
                                        high=list(df.high.values),
                                        low=list(df.low.values),
                                        volume=list(df.volume.values),
                                        time=list(df.time.dt.date.values),
                                        color=list(color.values)))
    
    

    # hover setting
    TOOLTIPS = [
            ("time", "@time{%F}"),
            ("adjusted close", "$@adjusted_close"),
            ("close", "$@close"),
            ("open", "$@open"),
            ("high", "$@high"),
            ("low", "$@low"),
            ("volume","@volume")]
    formatters={
        'time'      : 'datetime'}
    hover = HoverTool(tooltips=TOOLTIPS,formatters=formatters,mode='vline')
    
    # create figure
    p = figure(title='{} Candlestick'.format(stock_select.value),plot_height=400, 
               tools="crosshair,save,undo,xpan,xwheel_zoom,xbox_zoom,reset", 
               active_scroll='xwheel_zoom',
               x_axis_type="datetime")  
    p.add_tools(hover)
    


    p.line('time', 'close', alpha=0.2, line_width=1, color='navy', source=source)
    p.segment('time', 'high', 'time', 'low', line_width=1,color="black", source=source)
    p.segment('time', 'open', 'time', 'close', line_width=3, color='color', source=source)
    p.y_range = Range1d(min(source.data['close'])*0.3, max(source.data['close'])*1.05)

    
    p.extra_y_ranges = {"volumes": Range1d(start=min(source.data['volume'])/2, 
                                           end=max(source.data['volume'])*2)}
    p.add_layout(LinearAxis(y_range_name="volumes"), 'right')
    p.vbar('time', width=3,top='volume', color=choice(all_palettes['Set2'][8]),alpha=0.5, y_range_name="volumes",source=source)

    p.xaxis.axis_label = 'Time'
    
    # set data source
    _,_,_,institution_df=read_company(symbol_list[0])
    source_ins = ColumnDataSource(data=dict(organization=list(institution_df.organization.values),
                                            pctHeld=list(institution_df.pctHeld.values),
                                            position=list(institution_df.position.values),
                                            color=Set3[12][:len(institution_df)]))
    s1=figure(x_range=source_ins.data['organization'],plot_height=300,plot_width=700,title='Institution Ownership')
    s1.vbar(x='organization', top='position', width=0.8, color='color', source=source_ins)
    s1.xaxis.major_label_orientation = pi/7
    labels = LabelSet(x='organization', y='position', text='pctHeld', level='glyph',
              x_offset=-15, y_offset=-10, source=source_ins, render_mode='canvas',text_font_size="8pt")
    s1.add_layout(labels)
    # callback funtion for Select tool 'stock_select'
    def callback(attr,old,new):
        symbol=stock_select.value
        _,_,_,institution=read_company(symbol)

        if symbol=='S&P500':
            symbol='^GSPC'
        database=CassandraStorage(symbol)
        database.session.row_factory = pandas_factory
        database.session.default_fetch_size = None
        if symbol=='^GSPC':
            symbol='GSPC'
        query="SELECT * FROM {} WHERE time>'2015-01-01'  ALLOW FILTERING;".format(symbol+'_historical')
        rslt = database.session.execute(query, timeout=None)
        df = rslt._current_rows
    
        color=df.close>df.open
        color=color.replace(True,'green')
        color=color.replace(False,'red')
        
        # update source data 
        source.data=dict(close=list(df.close.values),
                                        adjusted_close=list(df.adjusted_close.values),
                                        open=list(df.open.values),
                                        high=list(df.high.values),
                                        low=list(df.low.values),
                                        volume=list(df.volume.values),
                                        time=list(df.time.dt.date.values),
                                        color=list(color.values))
        source_ins.data=dict(organization=list(institution.organization.values),
                                        pctHeld=list(institution.pctHeld.values),
                                        position=list(institution.position.values),
                                        color=Set3[12][:len(institution)])
        
        p.title.text=symbol+' Candlestick'
        p.y_range.start=min(source.data['close'])*0.3
        p.y_range.end=max(source.data['close'])*1.05
        p.extra_y_ranges['volumes'].start=min(source.data['volume'])/2.
        p.extra_y_ranges['volumes'].end=max(source.data['volume'])*2.

        s1.x_range.factors=source_ins.data['organization']
        update_summary(symbol)
        
    stock_select.on_change('value', callback)
    
    return p,stock_select,summaryText,financialText,s1

def stream_price():

def stream_price():
    
    # connect to s&p500's database
    plot_symbol='^GSPC'
    database=CassandraStorage(plot_symbol)
    database.session.row_factory = pandas_factory
    database.session.default_fetch_size = None
    
#    if datetime.datetime.now(timezone('US/Eastern')).time()  <  datetime.time(9,30):
#        query_time=str(datetime.datetime.now().date())
    
    
    last_trading_day= datetime.datetime.now(timezone(timeZone)).date()
    
    query="SELECT * FROM {} WHERE time>='{}'  ALLOW FILTERING;".format(plot_symbol[1:]+'_tick',last_trading_day)
    rslt = database.session.execute(query, timeout=None)
    df = rslt._current_rows

        # wrangle timezone (Cassandra will change datetime to UTC time)
    trans_time=pd.DatetimeIndex(pd.to_datetime(df.time,unit='ms')).tz_localize('GMT').tz_convert('US/Pacific').to_pydatetime()
    trans_time=[i.replace(tzinfo=None) for i in trans_time]
    source= ColumnDataSource()
    
    # hover setting
    TOOLTIPS = [
            ("time", "@time{%F %T}"),
            ("close", "$@close"),
            ("volume","@volume")]
    formatters={
        'time'      : 'datetime'}
    hover = HoverTool(tooltips=TOOLTIPS,formatters=formatters,mode='vline')

    # create plot
    p = figure(title='S&P500 Realtime Price',
                plot_height=200,  
                tools="crosshair,save,undo,xpan,xwheel_zoom,ybox_zoom,reset", 
                x_axis_type="datetime", 
                y_axis_location="left")
    p.add_tools(hover)
    p.x_range.follow = "end"
    p.x_range.follow_interval = 1000000
    p.x_range.range_padding = 0

    # during trading
    if len(df)>0 \
        and datetime.datetime.now(timezone(timeZone)).time() < datetime.time(16,0,0) \
        and datetime.datetime.now(timezone(timeZone)).time()>datetime.time(9,30,0):
        # init source data to those already stored in Cassandra dataase - '{}_tick', so that streaming plot will not start over after refreshing
        source= ColumnDataSource(dict(time=list(trans_time),  
                                       close=list(df.close.values),
                                       volume=list(df.volume.values)))
        p.y_range = Range1d(min(source.data['close'])/1.005, max(source.data['close'])*1.005)
        p.extra_y_ranges = {"volumes": Range1d(start=min(source.data['volume'])*0.5, 
                                               end=max(source.data['volume'])*2)}
    # no trading history or not during trading hour
    else:
        source= ColumnDataSource(dict(time=[],  
                                       close=[],
                                       volume=[]))
    
        p.y_range = Range1d(0,1e4)
        p.extra_y_ranges = {"volumes": Range1d(start=0, 
                                               end=1e10)}
        
    p.line(x='time', y='close', alpha=0.2, line_width=3, color='blue', source=source)    
    p.add_layout(LinearAxis(y_range_name="volumes"), 'right')    
    p.vbar('time', width=3,top='volume', color=choice(all_palettes['Set2'][8]),alpha=0.5, y_range_name="volumes",source=source)
    
    
    # get update data from a json file overwritter every ~18 seconds
    def _create_prices():
        with open(path+'cache/data.json','r') as f:
            dict_data = json.load(f)
        return float(dict_data['close']),dict_data['volume'],dict_data['time']
    
    # update function for stream plot
    def update():
        close,volume,time=_create_prices()
        new_data = dict(
            time=[datetime.datetime.strptime(time[:19], "%Y-%m-%d %H:%M:%S")],
            close=[close],
            volume=[volume]
        )
        #print(new_data)
        source.stream(new_data)
        #print ('update source data',str(time))
    return p,update

def make_lightcurve_figure_elements(lc, lc_source):
    """Make the lightcurve figure elements.

    Parameters
    ----------
    lc : LightCurve
        Lightcurve to be shown.
    lc_source : bokeh.plotting.ColumnDataSource
        Bokeh object that enables the visualization.

    Returns
    ----------
    fig : `bokeh.plotting.figure` instance
    step_renderer : GlyphRenderer
    vertical_line : Span
    """
    if lc.mission == 'K2':
        title = "Lightcurve for {} (K2 C{})".format(
            lc.label, lc.campaign)
    elif lc.mission == 'Kepler':
        title = "Lightcurve for {} (Kepler Q{})".format(
            lc.label, lc.quarter)
    elif lc.mission == 'TESS':
        title = "Lightcurve for {} (TESS Sec. {})".format(
            lc.label, lc.sector)
    else:
        title = "Lightcurve for target {}".format(lc.label)

    fig = figure(title=title, plot_height=340, plot_width=600,
                 tools="pan,wheel_zoom,box_zoom,tap,reset",
                 toolbar_location="below",
                 border_fill_color="whitesmoke")
    fig.title.offset = -10
    fig.yaxis.axis_label = 'Flux (e/s)'
    fig.xaxis.axis_label = 'Time (days)'
    try:
        if (lc.mission == 'K2') or (lc.mission == 'Kepler'):
            fig.xaxis.axis_label = 'Time - 2454833 (days)'
        elif lc.mission == 'TESS':
            fig.xaxis.axis_label = 'Time - 2457000 (days)'
    except AttributeError:  # no mission keyword available
      pass


    ylims = get_lightcurve_y_limits(lc_source)
    fig.y_range = Range1d(start=ylims[0], end=ylims[1])

    # Add step lines, circles, and hover-over tooltips
    fig.step('time', 'flux', line_width=1, color='gray',
             source=lc_source, nonselection_line_color='gray',
             nonselection_line_alpha=1.0)
    circ = fig.circle('time', 'flux', source=lc_source, fill_alpha=0.3, size=8,
                      line_color=None, selection_color="firebrick",
                      nonselection_fill_alpha=0.0,
                      nonselection_fill_color="grey",
                      nonselection_line_color=None,
                      nonselection_line_alpha=0.0,
                      fill_color=None, hover_fill_color="firebrick",
                      hover_alpha=0.9, hover_line_color="white")
    tooltips = [("Cadence", "@cadence"),
                ("Time ({})".format(lc.time_format.upper()),
                 "@time{0,0.000}"),
                ("Time (ISO)", "@time_iso"),
                ("Flux", "@flux"),
                ("Quality Code", "@quality_code"),
                ("Quality Flag", "@quality")]
    fig.add_tools(HoverTool(tooltips=tooltips, renderers=[circ],
                            mode='mouse', point_policy="snap_to_data"))

    # Vertical line to indicate the cadence
    vertical_line = Span(location=lc.time[0], dimension='height',
                         line_color='firebrick', line_width=4, line_alpha=0.5)
    fig.add_layout(vertical_line)

    return fig, vertical_line


def add_gaia_figure_elements(tpf, fig, magnitude_limit=18):

See More Examples