Plotting Interactive Bar Plot in Bokeh

Community Support
1

Hi Team,

I was trying to plot bar graph using bokeh and facing following issues.

  1. Interactivity is not there, I am using bokeh.charts and Bar library to create bar charts.

Below is the detail of the plot/function I am using:

def make_plot_promotype(source, title):
df=source.to_df()
bar = Bar(df,‘promotion_type’,values=‘value’,title=title)
return bar

Below is how I am calling this make_plot_promotype function

plot_sales_by_promo = make_plot_promotype(source_sales_by_promo, “Sales by Promotion Type”)

Now plotting this bar chart using

curdoc().add_root(HBox(plot_sales_by_promo))

The problem with the above code is that I am seeing two curves while running this application through bokeh server. One where I have called

plot_sales_by_promo = make_plot_promotype(source_sales_by_promo, “Sales by Promotion Type”)

and another is inside curdoc() function.

for implementing interactivity I am using call backs like this:

year_select.on_change(‘value’,update_plot)

where inside update_plot I am changing and update the source dataframe.

  1. Another problem with bokeh bar charts is that I want labels as strings on the X-axis but could not do this.

Appreciate your help on the same team.

Thanks,

Saurabh Heda

2

Somehow managed to plot Bar using rect but now the problem is related to the x axis labels. following is the code to plot the rectangle.

def make_plot_promotype(source, title):

  • label=[‘promo1’,‘promo2’,‘promo3’,‘promo4’]*

  • yr = DataRange1d(start=0, bounds=None)*

  • df=source.to_df()*

  • plot = Figure(plot_width=500, plot_height=400, toolbar_location=“above”,y_range=yr)*

  • plot.title = title*

  • plot.rect(x=‘promotion_type’,y=‘valuebytwo’,source=source, width =0.5, height = ‘value’, legend=title)*

  • plot.xaxis.major_label_orientation = np.pi/4 # radians, “horizontal”, “vertical”, “normal”*

  • return plot*

source is the columndatasource which is pandas dataframe. promotion_type, valuebytwo and value are the columns of the dataframe.

Above is the plot we get while x_range = label is not supplied but while figure is supplied with x_range = label getting the below curve.

plot = Figure(x_range=label,plot_width=500, plot_height=400, toolbar_location="above",y_range=yr)

···

On Tue, Sep 20, 2016 at 9:54 AM, Saurabh Heda [email protected] wrote:

Hi Team,

I was trying to plot bar graph using bokeh and facing following issues.

  1. Interactivity is not there, I am using bokeh.charts and Bar library to create bar charts.

Below is the detail of the plot/function I am using:

def make_plot_promotype(source, title):
df=source.to_df()
bar = Bar(df,‘promotion_type’,values=‘value’,title=title)
return bar

Below is how I am calling this make_plot_promotype function

plot_sales_by_promo = make_plot_promotype(source_sales_by_promo, “Sales by Promotion Type”)

Now plotting this bar chart using

curdoc().add_root(HBox(plot_sales_by_promo))

The problem with the above code is that I am seeing two curves while running this application through bokeh server. One where I have called

plot_sales_by_promo = make_plot_promotype(source_sales_by_promo, “Sales by Promotion Type”)

and another is inside curdoc() function.

for implementing interactivity I am using call backs like this:

year_select.on_change(‘value’,update_plot)

where inside update_plot I am changing and update the source dataframe.

  1. Another problem with bokeh bar charts is that I want labels as strings on the X-axis but could not do this.

Appreciate your help on the same team.

Thanks,

Saurabh Heda

–

You received this message because you are subscribed to the Google Groups “Bokeh Discussion - Public” group.

To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].

To post to this group, send email to [email protected].

To view this discussion on the web visit https://groups.google.com/a/continuum.io/d/msgid/bokeh/8e47f152-c056-4259-a026-220af56ed122%40continuum.io.

For more options, visit https://groups.google.com/a/continuum.io/d/optout.

–

Regards
Saurabh Heda
www.pg4research.blogspot.com
home.iitj.ac.in/~saurabh.heda

3

Can anybody help me out on this? Seems like bug to me.

···

On Thu, Sep 22, 2016 at 4:52 PM, Saurabh Heda [email protected] wrote:

Somehow managed to plot Bar using rect but now the problem is related to the x axis labels. following is the code to plot the rectangle.

def make_plot_promotype(source, title):

  • label=[‘promo1’,‘promo2’,‘promo3’,‘promo4’]*
  • yr = DataRange1d(start=0, bounds=None)*
  • df=source.to_df()*
  • plot = Figure(plot_width=500, plot_height=400, toolbar_location=“above”,y_range=yr)*
  • plot.title = title*
  • plot.rect(x=‘promotion_type’,y=‘valuebytwo’,source=source, width =0.5, height = ‘value’, legend=title)*
  • plot.xaxis.major_label_orientation = np.pi/4 # radians, “horizontal”, “vertical”, “normal”*
  • return plot*

source is the columndatasource which is pandas dataframe. promotion_type, valuebytwo and value are the columns of the dataframe.

Above is the plot we get while x_range = label is not supplied but while figure is supplied with x_range = label getting the below curve.

plot = Figure(x_range=label,plot_width=500, plot_height=400, toolbar_location=“above”,y_range=yr)

–

On Tue, Sep 20, 2016 at 9:54 AM, Saurabh Heda [email protected] wrote:

Hi Team,

I was trying to plot bar graph using bokeh and facing following issues.

  1. Interactivity is not there, I am using bokeh.charts and Bar library to create bar charts.

Below is the detail of the plot/function I am using:

def make_plot_promotype(source, title):
df=source.to_df()
bar = Bar(df,‘promotion_type’,values=‘value’,title=title)
return bar

Below is how I am calling this make_plot_promotype function

plot_sales_by_promo = make_plot_promotype(source_sales_by_promo, “Sales by Promotion Type”)

Now plotting this bar chart using

curdoc().add_root(HBox(plot_sales_by_promo))

The problem with the above code is that I am seeing two curves while running this application through bokeh server. One where I have called

plot_sales_by_promo = make_plot_promotype(source_sales_by_promo, “Sales by Promotion Type”)

and another is inside curdoc() function.

for implementing interactivity I am using call backs like this:

year_select.on_change(‘value’,update_plot)

where inside update_plot I am changing and update the source dataframe.

  1. Another problem with bokeh bar charts is that I want labels as strings on the X-axis but could not do this.

Appreciate your help on the same team.

Thanks,

Saurabh Heda

–

You received this message because you are subscribed to the Google Groups “Bokeh Discussion - Public” group.

To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].

To post to this group, send email to [email protected].

To view this discussion on the web visit https://groups.google.com/a/continuum.io/d/msgid/bokeh/8e47f152-c056-4259-a026-220af56ed122%40continuum.io.

For more options, visit https://groups.google.com/a/continuum.io/d/optout.

Regards
Saurabh Heda
www.pg4research.blogspot.com
home.iitj.ac.in/~saurabh.heda

–

Regards
Saurabh Heda
www.pg4research.blogspot.com
home.iitj.ac.in/~saurabh.heda

4

Can you print and share the value of source.data[‘promitiin_type’] for us?

Thanks,

Bryan

···

On Thu, Sep 22, 2016 at 4:52 PM, Saurabh Heda [email protected] wrote:

Somehow managed to plot Bar using rect but now the problem is related to the x axis labels. following is the code to plot the rectangle.

def make_plot_promotype(source, title):

  • label=[‘promo1’,‘promo2’,‘promo3’,‘promo4’]*
  • yr = DataRange1d(start=0, bounds=None)*
  • df=source.to_df()*
  • plot = Figure(plot_width=500, plot_height=400, toolbar_location=“above”,y_range=yr)*
  • plot.title = title*
  • plot.rect(x=‘promotion_type’,y=‘valuebytwo’,source=source, width =0.5, height = ‘value’, legend=title)*
  • plot.xaxis.major_label_orientation = np.pi/4 # radians, “horizontal”, “vertical”, “normal”*
  • return plot*

source is the columndatasource which is pandas dataframe. promotion_type, valuebytwo and value are the columns of the dataframe.

<image.png>

Above is the plot we get while x_range = label is not supplied but while figure is supplied with x_range = label getting the below curve.

plot = Figure(x_range=label,plot_width=500, plot_height=400, toolbar_location=“above”,y_range=yr)

<image.png>

–

On Tue, Sep 20, 2016 at 9:54 AM, Saurabh Heda [email protected] wrote:

Hi Team,

I was trying to plot bar graph using bokeh and facing following issues.

  1. Interactivity is not there, I am using bokeh.charts and Bar library to create bar charts.

Below is the detail of the plot/function I am using:

def make_plot_promotype(source, title):
df=source.to_df()
bar = Bar(df,‘promotion_type’,values=‘value’,title=title)
return bar

Below is how I am calling this make_plot_promotype function

plot_sales_by_promo = make_plot_promotype(source_sales_by_promo, “Sales by Promotion Type”)

Now plotting this bar chart using

curdoc().add_root(HBox(plot_sales_by_promo))

The problem with the above code is that I am seeing two curves while running this application through bokeh server. One where I have called

plot_sales_by_promo = make_plot_promotype(source_sales_by_promo, “Sales by Promotion Type”)

and another is inside curdoc() function.

for implementing interactivity I am using call backs like this:

year_select.on_change(‘value’,update_plot)

where inside update_plot I am changing and update the source dataframe.

  1. Another problem with bokeh bar charts is that I want labels as strings on the X-axis but could not do this.

Appreciate your help on the same team.

Thanks,

Saurabh Heda

–

You received this message because you are subscribed to the Google Groups “Bokeh Discussion - Public” group.

To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].

To post to this group, send email to [email protected].

To view this discussion on the web visit https://groups.google.com/a/continuum.io/d/msgid/bokeh/8e47f152-c056-4259-a026-220af56ed122%40continuum.io.

For more options, visit https://groups.google.com/a/continuum.io/d/optout.

Regards
Saurabh Heda
www.pg4research.blogspot.com
home.iitj.ac.in/~saurabh.heda

–

Regards
Saurabh Heda
www.pg4research.blogspot.com
home.iitj.ac.in/~saurabh.heda