Allow Bars to be plotted on continuous axes #6145

philippjfr · 2024-03-06T23:43:38Z

About 10 years late, but better than never.

hv.Bars(range(1, 10))

df.time.dt.date.value_counts().sort_index().iloc[:30].hvplot.bar(title='Daily Earthquakes')

MarcSkovMadsen · 2024-03-09T17:11:15Z

I've tested it out.

bokeh backend: width of bars should be fixed. Right now they are dynamic. Zoom, Hover, Layouts etc. seem to work. Nice.
matplotlib backend: not showing/ working
plotly backend: not showing/ working

Getting Bokeh backend working is a big improvement for me and what I will be using most of the time. Getting Plotly backend working would be really nice and I see many users in business preferring that backend.

test-bars-with-axis-series-ezgif.com-video-speed.mp4

Code

import holoviews as hv
import panel as pn
import pandas as pd
import pandas as pd
import numpy as np
from datetime import date
import param
import math
import hvplot.pandas

pn.extension("plotly", sizing_mode="stretch_width")


min_price = 90
max_price = 110

ACCENT = "#3d5a80"
POS = "#98C1D9"
NEG = "#ee6c4d"

PERIODS = {
    "MS": (date(2022, 1, 1), date(2024, 12, 1)),
    "ME": (date(2022, 1, 1), date(2024, 12, 1)),
    "D": (date(2022, 1, 1), date(2022, 7, 1)),
    "B": (date(2022, 1, 1), date(2022, 7, 1)),
    "h": (date(2022, 1, 1), date(2022, 1, 8)),
}
OPTIONS={
    "bokeh": {
        "line_width": 5
    },
    "matplotlib": {
        "linewidth": 5
    },
    "plotly": {
        "line_width": 5
    }
}
def _get_label(text: str):
    return pn.pane.Markdown(text, margin=(-20, 5, -20, 5))

class PricePlotter(pn.viewable.Viewer):
    frequency = param.Selector(default="B", objects=list(PERIODS))
    start_date = param.CalendarDate(date(2022, 1, 1))
    end_date = param.CalendarDate(date(2022, 4, 1))
    lib = param.Selector(default="hvPlot", objects=["hvPlot", "HoloViews"])
    backend = param.Selector(default="bokeh", objects=["bokeh", "matplotlib", "plotly"])

    def __init__(self, **params):
        super().__init__(**params)

        self.settings = pn.Column(
            _get_label("Frequency"),
            pn.widgets.RadioButtonGroup.from_param(
                self.param.frequency, button_type="primary", button_style="outline"
            ),
            self.param.start_date,
            self.param.end_date,
            _get_label("Library"),
            pn.widgets.RadioButtonGroup.from_param(
                self.param.lib, button_type="primary", button_style="outline"
            ),
            _get_label("Backend"),
            pn.widgets.RadioButtonGroup.from_param(
                self.param.backend, button_type="primary", button_style="outline"
            )
        )
        self._layout = pn.Row(self.settings, self.get_plot)

    def __panel__(self):
        return self._layout

    @param.depends("frequency", watch=True, on_init=True)
    def _update_period(self):
        start, end = PERIODS[self.frequency]
        self.param.update(start_date=start, end_date=end)

    @param.depends("backend", watch=True, on_init=True)
    def _update_backend(self):
        hv.extension(self.backend)

    def get_prices(self):
        dates = pd.date_range(
            start=self.start_date, end=self.end_date, freq=self.frequency
        )
        num_values = len(dates)
        values = np.random.uniform(min_price, max_price, size=num_values)
        df = pd.DataFrame({"date": dates, "price": values})
        df["diff"] = df["price"].diff()
        df["color"] = (df["diff"] >= 0).map({True: "blue", False: "red"})
        df = df.dropna()
        return df

    def get_plot(self):

        df = self.get_prices()
        df_pos = df[df["diff"] >= 0]
        df_neg = df[df["diff"] <= 0]

        max_price = df["price"].max()
        max_y = math.ceil(max_price / 10) * 10 + 10
        options=OPTIONS[self.backend]
        if self.lib == "HoloViews":
            abs_plot = hv.Curve(df, kdims="date", vdims="price").opts(
                responsive=True,
                height=500,
                color=ACCENT,
                ylim=(0, max_y),
                tools=["hover"],
                **options
            )
            rel_plot_pos = hv.Bars(
                df_pos,
                kdims="date",
                vdims="diff",
            ).opts(responsive=True, height=200, color=POS, tools=["hover"])
            rel_plot_neg = hv.Bars(
                df_neg,
                kdims="date",
                vdims="diff",
            ).opts(responsive=True, height=200, color=NEG, tools=["hover"])
        else:
            abs_plot = df.hvplot.line(
                x="date",
                y="price",
                responsive=True,
                height=500,
                color=ACCENT,
                ylim=(0, max_y),
                tools=["hover"],
                xlabel="Date",
                ylabel="Price",
                **options
            )
            rel_plot_pos = df_pos.hvplot.bar(
                x="date",
                y="diff",
                responsive=True,
                height=200,
                color=POS,
                tools=["hover"],
                xlabel="Date",
                ylabel="Diff",
            )
            rel_plot_neg = df_neg.hvplot.bar(
                x="date",
                y="diff",
                responsive=True,
                height=200,
                color=NEG,
                tools=["hover"],
                xlabel="Date",
                ylabel="Diff",
            )
        return (abs_plot + rel_plot_pos * rel_plot_neg).cols(1)


plotter = PricePlotter()

pn.template.FastListTemplate(
    title="Bars Test App", sidebar=[plotter.settings], main=[plotter.get_plot], main_layout=None, accent=ACCENT
).servable()

MarcSkovMadsen · 2024-03-20T16:42:21Z

Matplotlib Analysis

This is how Matplotlib does it.

App

matplotlib-bars.mp4

Code

import panel as pn
import pandas as pd
import pandas as pd
import numpy as np
from datetime import date
import param
import matplotlib.pyplot as plt

pn.extension("plotly", sizing_mode="stretch_width")


min_price = 90
max_price = 110

ACCENT = "#3d5a80"
POS = "#98C1D9"
NEG = "#ee6c4d"

PERIODS = {
    "MS": (date(2022, 1, 1), date(2024, 12, 1)),
    "ME": (date(2022, 1, 1), date(2024, 12, 1)),
    "D": (date(2022, 1, 1), date(2022, 7, 1)),
    "B": (date(2022, 1, 1), date(2022, 7, 1)),
    "h": (date(2022, 1, 1), date(2022, 1, 8)),
}
OPTIONS={
    "bokeh": {
        "line_width": 5
    },
    "matplotlib": {
        "linewidth": 5
    },
    "plotly": {
        "line_width": 5
    }
}
def _get_label(text: str):
    return pn.pane.Markdown(text, margin=(-20, 5, -20, 5))

class PricePlotter(pn.viewable.Viewer):
    frequency = param.Selector(default="B", objects=list(PERIODS))
    start_date = param.CalendarDate(date(2022, 1, 1))
    end_date = param.CalendarDate(date(2022, 4, 1))
    
    def __init__(self, **params):
        super().__init__(**params)

        self.settings = pn.Column(
            _get_label("Frequency"),
            pn.widgets.RadioButtonGroup.from_param(
                self.param.frequency, button_type="primary", button_style="outline"
            ),
            self.param.start_date,
            self.param.end_date,
        )
        self._layout = pn.Row(self.settings, self.get_plot)

    def __panel__(self):
        return self._layout

    @param.depends("frequency", watch=True, on_init=True)
    def _update_period(self):
        start, end = PERIODS[self.frequency]
        self.param.update(start_date=start, end_date=end)

    def get_prices(self):
        dates = pd.date_range(
            start=self.start_date, end=self.end_date, freq=self.frequency
        )
        num_values = len(dates)
        values = np.random.uniform(min_price, max_price, size=num_values)
        df = pd.DataFrame({"date": dates, "price": values})
        df["diff"] = df["price"].diff()
        df["color"] = (df["diff"] >= 0).map({True: "blue", False: "red"})
        df = df.dropna()
        return df

    def get_plot(self):

        df = self.get_prices()
        df_pos = df[df["diff"] >= 0]
        df_neg = df[df["diff"] <= 0]

        max_price = df["price"].max()

        fig1 = plt.figure(figsize=(10, 3))
        plt.plot(df["date"], df["price"], color=ACCENT, label="Price")
        plt.xlabel("Date")
        plt.ylabel("Price")
        plt.title("Price Over Time")
        plt.legend()
        plt.close(fig1)

        # Plotting the bar plots for positive and negative differences
        fig2 = plt.figure(figsize=(10, 3))

        # Positive differences
        plt.bar(df_pos["date"], df_pos["diff"], color=POS, label="Positive Diff")
        # Negative differences
        plt.bar(df_neg["date"], df_neg["diff"], color=NEG, label="Negative Diff")

        plt.xlabel("Date")
        plt.ylabel("Diff")
        plt.title("Relative Differences Over Time")
        plt.legend()
        plt.close(fig2) # CLOSE THE FIGURE!
        return pn.Column(
            pn.pane.Matplotlib(fig1, tight=True), 
            pn.pane.Matplotlib(fig2, tight=True),
        )


plotter = PricePlotter()

pn.template.FastListTemplate(
    title="Bars Test App", sidebar=[plotter.settings], main=[plotter.get_plot], main_layout=None, accent=ACCENT
).servable()

Matplotlib Bars Use a Fixed Default Width

Matplotlib uses a fixed, default width of 0.8.

See https://github.com/matplotlib/matplotlib/blob/df148eea635d3c6855565681f3275db6dd2be436/lib/matplotlib/axes/_axes.py#L2326

MarcSkovMadsen · 2024-03-20T17:05:50Z

Pandas `.plot` analysis

This is how Pandas .plot does it.

App

pandas-plot-bar-ezgif.com-video-speed.mp4

Code

import panel as pn
import pandas as pd
import pandas as pd
import numpy as np
from datetime import date
import param
import matplotlib.pyplot as plt

pn.extension("plotly", sizing_mode="stretch_width")


min_price = 90
max_price = 110

ACCENT = "#3d5a80"
POS = "#98C1D9"
NEG = "#ee6c4d"

PERIODS = {
    "MS": (date(2022, 1, 1), date(2024, 12, 1)),
    "ME": (date(2022, 1, 1), date(2024, 12, 1)),
    "D": (date(2022, 1, 1), date(2022, 7, 1)),
    "B": (date(2022, 1, 1), date(2022, 7, 1)),
    "h": (date(2022, 1, 1), date(2022, 1, 8)),
}
OPTIONS={
    "bokeh": {
        "line_width": 5
    },
    "matplotlib": {
        "linewidth": 5
    },
    "plotly": {
        "line_width": 5
    }
}
def _get_label(text: str):
    return pn.pane.Markdown(text, margin=(-20, 5, -20, 5))

class PricePlotter(pn.viewable.Viewer):
    frequency = param.Selector(default="B", objects=list(PERIODS))
    start_date = param.CalendarDate(date(2022, 1, 1))
    end_date = param.CalendarDate(date(2022, 4, 1))
    
    def __init__(self, **params):
        super().__init__(**params)

        self.settings = pn.Column(
            _get_label("Frequency"),
            pn.widgets.RadioButtonGroup.from_param(
                self.param.frequency, button_type="primary", button_style="outline"
            ),
            self.param.start_date,
            self.param.end_date,
        )
        self._layout = pn.Row(self.settings, self.get_plot)

    def __panel__(self):
        return self._layout

    @param.depends("frequency", watch=True, on_init=True)
    def _update_period(self):
        start, end = PERIODS[self.frequency]
        self.param.update(start_date=start, end_date=end)

    def get_prices(self):
        dates = pd.date_range(
            start=self.start_date, end=self.end_date, freq=self.frequency
        )
        num_values = len(dates)
        values = np.random.uniform(min_price, max_price, size=num_values)
        df = pd.DataFrame({"date": dates, "price": values})
        df["diff"] = df["price"].diff()
        df["color"] = (df["diff"] >= 0).map({True: "blue", False: "red"})
        df = df.dropna()
        return df

    def get_plot(self):

        df = self.get_prices()
        df_pos = df[df["diff"] >= 0]
        df_neg = df[df["diff"] <= 0]

        max_price = df["price"].max()

        fig1 = plt.figure(figsize=(10, 3))
        plt.plot(df["date"], df["price"], color=ACCENT, label="Price")
        plt.xlabel("Date")
        plt.ylabel("Price")
        plt.title("Price Over Time")
        plt.legend()
        plt.close(fig1)

        # Plotting the bar plots for positive and negative differences
        df_pos['date'] = pd.to_datetime(df_pos['date'])
        df_neg['date'] = pd.to_datetime(df_neg['date'])

        # Set 'date' column as index for easy plotting
        df_pos.set_index('date', inplace=True)
        df_neg.set_index('date', inplace=True)

        # Concatenate positive and negative differences into a single DataFrame
        df_combined = pd.concat([df_pos['diff'], df_neg['diff']], axis=1)
        df_combined.columns = ['Positive Diff', 'Negative Diff']

        # Plot using pandas plot
        fig2 = plt.figure(figsize=(10, 3))
        ax = fig2.add_subplot(111)
        ax = df_combined.plot.bar(figsize=(10, 3), color=[POS, NEG], ax=ax, xlabel="Date", ylabel="Diff", title="Relative Differences Over Time")
        ax.legend(["Positive Diff", "Negative Diff"])
        plt.close(fig2) # CLOSE THE FIGURE!
        return pn.Column(
            pn.pane.Matplotlib(fig1, tight=True), 
            pn.pane.Matplotlib(fig2, tight=True),
        )


plotter = PricePlotter()

pn.template.FastListTemplate(
    title="Bars Test App", sidebar=[plotter.settings], main=[plotter.get_plot], main_layout=None, accent=ACCENT
).servable()

Implementation

The implemention is here. It looks to me as a default, fixed width of 0.5 is used ?? In practice I don't think that is what I see??

https://github.com/pandas-dev/pandas/blob/114a84d8a0eee8fb93a4d2d701a2a6e62ebcf6d2/pandas/plotting/_matplotlib/core.py#L1803

MarcSkovMadsen · 2024-03-20T17:56:23Z

Plotly Analysis

Plotly adjust the bar width based on the input. Clearly the best behaviour.

App

plotly-plot.mp4

Code

import panel as pn
import pandas as pd
import pandas as pd
import numpy as np
from datetime import date
import param
import plotly.express as px
from plotly.subplots import make_subplots

pn.extension("plotly", sizing_mode="stretch_width")


min_price = 90
max_price = 110

ACCENT = "#3d5a80"
POS = "#98C1D9"
NEG = "#ee6c4d"

PERIODS = {
    "MS": (date(2022, 1, 1), date(2024, 12, 1)),
    "ME": (date(2022, 1, 1), date(2024, 12, 1)),
    "D": (date(2022, 1, 1), date(2022, 7, 1)),
    "B": (date(2022, 1, 1), date(2022, 7, 1)),
    "h": (date(2022, 1, 1), date(2022, 1, 8)),
}
OPTIONS={
    "bokeh": {
        "line_width": 5
    },
    "matplotlib": {
        "linewidth": 5
    },
    "plotly": {
        "line_width": 5
    }
}
def _get_label(text: str):
    return pn.pane.Markdown(text, margin=(-20, 5, -20, 5))

class PricePlotter(pn.viewable.Viewer):
    frequency = param.Selector(default="B", objects=list(PERIODS))
    start_date = param.CalendarDate(date(2022, 1, 1))
    end_date = param.CalendarDate(date(2022, 4, 1))
    
    def __init__(self, **params):
        super().__init__(**params)

        self.settings = pn.Column(
            _get_label("Frequency"),
            pn.widgets.RadioButtonGroup.from_param(
                self.param.frequency, button_type="primary", button_style="outline"
            ),
            self.param.start_date,
            self.param.end_date,
        )
        self._layout = pn.Row(self.settings, self.get_plot)

    def __panel__(self):
        return self._layout

    @param.depends("frequency", watch=True, on_init=True)
    def _update_period(self):
        start, end = PERIODS[self.frequency]
        self.param.update(start_date=start, end_date=end)

    def get_prices(self):
        dates = pd.date_range(
            start=self.start_date, end=self.end_date, freq=self.frequency
        )
        num_values = len(dates)
        values = np.random.uniform(min_price, max_price, size=num_values)
        df = pd.DataFrame({"date": dates, "price": values})
        df["diff"] = df["price"].diff()
        df["color"] = (df["diff"] >= 0).map({True: "pos", False: "neg"})
        df = df.dropna()
        return df

    def get_plot(self):

        df = self.get_prices()
        abs_plot = px.line(
            df,
            x="date",
            y="price",
            height=400,
            color_discrete_sequence=[ACCENT],
            labels={"date": "Date", "price": "Price"},
            title="Absolute Price Over Time",
        )

        rel_plot_pos = px.bar(
            df,
            x="date",
            y="diff",
            height=400,
            color="color",
            color_discrete_sequence=[POS, NEG],
            labels={"date": "Date", "diff": "Diff"},
            title="Positive Relative Differences Over Time",
        )

        fig = make_subplots(rows=3, cols=1, shared_xaxes=True, vertical_spacing=0.03)
        # Add plots to subplots
        fig.add_trace(abs_plot.data[0], row=1, col=1)
        fig.add_trace(rel_plot_pos.data[0], row=2, col=1)
        fig.add_trace(rel_plot_pos.data[1], row=2, col=1)
        
        # Update layout
        fig.update_layout(
            height=800,
            title_text="Combined Plots",
            showlegend=False  # Assuming you don't want individual legends
        )
        return fig


plotter = PricePlotter()

pn.template.FastListTemplate(
    title="Bars Test App", sidebar=[plotter.settings], main=[plotter.get_plot], main_layout=None, accent=ACCENT
).servable()

Implementation Details

I've not been able to find the calculation of the bar width. I believe its done on Javascript side

hoxbro · 2024-03-20T18:18:49Z

I assume it was a mistake you closed this?

MarcSkovMadsen · 2024-03-20T19:16:58Z

Reflection on calculation of bar width.

Could the bar width be calculated by

sorting the timeseries
Calculating the length of all steps
Taking the minimum of all steps.
Adjusting with a factor 0.8

?

MarcSkovMadsen · 2024-03-20T19:30:25Z

Test of Bar Width formula

If you change the bar width calculation from

xdiff = np.diff(xvals).astype('timedelta64[ms]').astype(np.int32) * width * 0.5

to

xdiff = np.diff(xvals).astype('timedelta64[ms]').astype(np.int32) * width * 0.5
min_value = np.min(xdiff)
xdiff = np.full_like(xdiff, min_value)

It looks like below which looks correct to me.

bar-width-formula.mp4

I would recommend experimenting with the width factor. Probably using 0.8 instead of 0.5.

TheoMathurin · 2024-03-26T08:29:10Z

This is a very welcome addition!

droumis · 2024-03-27T16:41:34Z

Similarly, it would be great to be able to use distribution models like BoxWhisker and Violin on a datetime or other continuous x axis. For instance, if one has a distribution of continuous values grouped per day

codecov · 2024-04-22T20:34:59Z

Codecov Report

Attention: Patch coverage is 99.15612% with 2 lines in your changes are missing coverage. Please review.

Project coverage is 88.38%. Comparing base (b1fe1cf) to head (d1a9ca7).
Report is 10 commits behind head on main.

Files	Patch %	Lines
holoviews/plotting/mpl/chart.py	94.44%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #6145      +/-   ##
==========================================
+ Coverage   88.22%   88.38%   +0.16%     
==========================================
  Files         321      322       +1     
  Lines       67285    67559     +274     
==========================================
+ Hits        59361    59715     +354     
+ Misses       7924     7844      -80

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

ahuang11 · 2024-04-22T20:38:14Z

Okay, MPL partially fixed

import holoviews as hv
import pandas as pd
import matplotlib.pyplot as plt
# hv.extension("matplotlib")

df = pd.DataFrame({"x": [1, 2, 3, 4, 5], "y": [10, 20, 30, 40, 50]})

fig = plt.figure()
ax = fig.add_subplot(111)
ax.bar(df["x"], df["y"])

hv.extension()

display(hv.Bars(df, ["x"], ["y"]).opts(bar_padding=0.1, align="center"))
hv.Bars(df, ["x"], ["y"]).opts(bar_padding=0.1, align="edge")

Reference:

Categories:

import numpy as np
import holoviews as hv
hv.extension('matplotlib')

data = [('one',8),('two', 10), ('three', 16), ('four', 8), ('five', 4), ('six', 1)]

bars = hv.Bars(data, hv.Dimension('Car occupants'), 'Count')

occupants = hv.Dimension('Car occupants', values=['three', 'two', 'four', 'one', 'five', 'six'])

# or using .redim.values(**{'Car Occupants': ['three', 'two', 'four', 'one', 'five', 'six']})

hv.Bars(data, occupants, 'Count')

Groups:

samples = 100

pets = ['Cat', 'Dog', 'Hamster', 'Rabbit']
genders = ['Female', 'Male', "A", "B"]

np.random.seed(100)
pets_sample = np.random.choice(pets, samples)
gender_sample = np.random.choice(genders, samples)

bars = hv.Bars((pets_sample, gender_sample, np.ones(samples)), ['Pets', 'Gender']).aggregate(function=np.sum)

bars.opts(fig_size=300, aspect=2, show_legend=True)

Stacked:

Note for self; things to test:

ahuang11 · 2024-04-23T02:50:53Z

hv.Bars matplotlib backend is in a much better state now, with tests added!

Also, made continuous axes actually continuous, matching matplotlib behavior

import numpy as np
import pandas as pd
import holoviews as hv
import matplotlib.pyplot as plt
from holoviews.element import Bars
from holoviews.plotting.mpl import Store
import matplotlib.dates as mdates

hv.extension("matplotlib")

mpl_renderer = Store.renderers["matplotlib"]

hv.Bars(([0, 1, 2, 10], [10, 20, 30, 40]))

ahuang11 · 2024-05-02T00:43:09Z

Make plotly continuous

import matplotlib.pyplot as plt
import holoviews as hv
import xarray as xr

hv.extension("plotly")

ds = xr.tutorial.open_dataset("air_temperature")
df = ds.resample(time="1MS").mean().mean(["lat", "lon"]).to_dataframe()
df

bars = hv.Bars(df, "time", "air")
bars

hoxbro

I have reviewed the code and left some comments.

I haven't yet tried out the code itself.

examples/reference/elements/bokeh/Bars.ipynb

holoviews/plotting/bokeh/chart.py

holoviews/plotting/plotly/chart.py

holoviews/tests/plotting/matplotlib/test.ipynb

holoviews/tests/plotting/matplotlib/test_barplot.py

holoviews/tests/plotting/plotly/test_barplot.py

Co-authored-by: Simon Høxbro Hansen <simon.hansen@me.com>

holoviews/plotting/bokeh/chart.py

holoviews/plotting/mpl/chart.py

hoxbro

Enumerate is not needed anymore, so I removed it.

ahuang11 · 2024-05-15T17:16:18Z

Strange, but glad it works!

philippjfr · 2024-05-17T11:07:54Z

Okay, I reviewed again and it looks good to me.

github-actions · 2024-10-23T10:27:47Z

This pull request has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.

Allow Bars to be plotted on continuous axes

971bcee

philippjfr mentioned this pull request Mar 6, 2024

Bar Chart x axis labels not formatted holoviz/hvplot#1282

Open

1 task

MarcSkovMadsen closed this Mar 20, 2024

hoxbro reopened this Mar 20, 2024

Fix matplotlib bars calculations

8b180a9

Add mini offset

a197c09

ahuang11 added the type: enhancement Minor feature or improvement to an existing feature label Apr 23, 2024

ahuang11 added 2 commits April 22, 2024 19:07

Fix poor logic and make actually continuous

96ce1c8

add tests

d20dbbc

ahuang11 added 2 commits April 27, 2024 20:07

Fix bokeh backend and add tests

783e56c

add docs

7023a5c

ahuang11 mentioned this pull request Apr 30, 2024

hist with logx produces buckets in non-logarithmic scale holoviz/hvplot#1313

Open

MarcSkovMadsen and others added 4 commits April 30, 2024 19:33

little iteration

ff1348b

rename

89660e8

fix group bars in mpl

baefbbd

make plotly continuous

62ce288

fix plotly tests and fix mpl ranges

63b9d92

Merge branch 'main' into continuous_bars

d3228b5

hoxbro requested changes May 3, 2024

View reviewed changes

ahuang11 and others added 3 commits May 6, 2024 06:43

address comments

030865d

remove version py

36334d0

Update examples/reference/elements/bokeh/Bars.ipynb

f816c13

Co-authored-by: Simon Høxbro Hansen <simon.hansen@me.com>

ahuang11 requested a review from hoxbro May 6, 2024 14:23

philippjfr commented May 6, 2024

View reviewed changes

holoviews/plotting/bokeh/chart.py Outdated Show resolved Hide resolved

ahuang11 added 3 commits May 6, 2024 08:51

rm width from data

b0eab97

rm width from data

5ce29b0

fix tests

be03d0c

philippjfr commented May 8, 2024

View reviewed changes

holoviews/plotting/bokeh/chart.py Outdated Show resolved Hide resolved

philippjfr commented May 8, 2024

View reviewed changes

holoviews/plotting/bokeh/chart.py Outdated Show resolved Hide resolved

ahuang11 added 2 commits May 8, 2024 06:00

address comments

bc5881d

fix tests

685bee9

hoxbro mentioned this pull request May 14, 2024

hv.Bars with more than one kdim randomly generates incorrect x-ticks in some cases (matplotlib backend) #6231

Closed

1 task

hoxbro linked an issue May 14, 2024 that may be closed by this pull request

hv.Bars with more than one kdim randomly generates incorrect x-ticks in some cases (matplotlib backend) #6231

Closed

1 task

ahuang11 reviewed May 14, 2024

View reviewed changes

holoviews/plotting/mpl/chart.py Outdated Show resolved Hide resolved

ahuang11 and others added 3 commits May 14, 2024 11:54

Update holoviews/plotting/mpl/chart.py

2b7c88a

fix test

9b8f0b7

Try to remove enumerate

d1a9ca7

hoxbro approved these changes May 15, 2024

View reviewed changes

hoxbro added this to the 1.19.0 milestone May 17, 2024

philippjfr merged commit ee20d23 into main May 17, 2024
13 checks passed

philippjfr deleted the continuous_bars branch May 17, 2024 11:10

philippjfr mentioned this pull request May 17, 2024

Align Bars with annotations on matplotlib #5024

Closed

TheoMathurin mentioned this pull request Jun 18, 2024

Stacked bars not rendered correctly with datetime axis #6288

Closed

github-actions bot locked as resolved and limited conversation to collaborators Oct 23, 2024

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Allow Bars to be plotted on continuous axes #6145

Allow Bars to be plotted on continuous axes #6145

philippjfr commented Mar 6, 2024

MarcSkovMadsen commented Mar 9, 2024 •

edited

Loading

MarcSkovMadsen commented Mar 20, 2024 •

edited

Loading

MarcSkovMadsen commented Mar 20, 2024 •

edited

Loading

MarcSkovMadsen commented Mar 20, 2024 •

edited

Loading

hoxbro commented Mar 20, 2024

MarcSkovMadsen commented Mar 20, 2024 •

edited

Loading

MarcSkovMadsen commented Mar 20, 2024 •

edited

Loading

TheoMathurin commented Mar 26, 2024

droumis commented Mar 27, 2024

codecov bot commented Apr 22, 2024 •

edited

Loading

ahuang11 commented Apr 22, 2024

ahuang11 commented Apr 23, 2024 •

edited

Loading

ahuang11 commented May 2, 2024

hoxbro left a comment

hoxbro left a comment

ahuang11 commented May 15, 2024

philippjfr commented May 17, 2024

github-actions bot commented Oct 23, 2024

Allow Bars to be plotted on continuous axes #6145

Allow Bars to be plotted on continuous axes #6145

Conversation

philippjfr commented Mar 6, 2024

MarcSkovMadsen commented Mar 9, 2024 • edited Loading

MarcSkovMadsen commented Mar 20, 2024 • edited Loading

Matplotlib Analysis

App

Matplotlib Bars Use a Fixed Default Width

MarcSkovMadsen commented Mar 20, 2024 • edited Loading

Pandas .plot analysis

App

Implementation

MarcSkovMadsen commented Mar 20, 2024 • edited Loading

Plotly Analysis

App

Implementation Details

hoxbro commented Mar 20, 2024

MarcSkovMadsen commented Mar 20, 2024 • edited Loading

Reflection on calculation of bar width.

MarcSkovMadsen commented Mar 20, 2024 • edited Loading

Test of Bar Width formula

TheoMathurin commented Mar 26, 2024

droumis commented Mar 27, 2024

codecov bot commented Apr 22, 2024 • edited Loading

Codecov Report

ahuang11 commented Apr 22, 2024

ahuang11 commented Apr 23, 2024 • edited Loading

ahuang11 commented May 2, 2024

hoxbro left a comment

Choose a reason for hiding this comment

hoxbro left a comment

Choose a reason for hiding this comment

ahuang11 commented May 15, 2024

philippjfr commented May 17, 2024

github-actions bot commented Oct 23, 2024

MarcSkovMadsen commented Mar 9, 2024 •

edited

Loading

MarcSkovMadsen commented Mar 20, 2024 •

edited

Loading

MarcSkovMadsen commented Mar 20, 2024 •

edited

Loading

Pandas `.plot` analysis

MarcSkovMadsen commented Mar 20, 2024 •

edited

Loading

MarcSkovMadsen commented Mar 20, 2024 •

edited

Loading

MarcSkovMadsen commented Mar 20, 2024 •

edited

Loading

codecov bot commented Apr 22, 2024 •

edited

Loading

ahuang11 commented Apr 23, 2024 •

edited

Loading