How do i set x axis to value of pandas df?

Community Support
1

I’m plotting OHLCV values with bokeh and having hard time figuring out how to set x axis to corresponding “open_time” from panda df.

This is how i plot

p = figure(plot_width=900, plot_height=400)#, x_axis_type='datetime')
    p.segment('bar', 'high', 'bar', 'low', source=src, line_width=1, color='black')  # plot the wicks
    p.vbar('bar', 0.7, 'close', 'open', source=src, line_color='white', fill_color='colors', )  # plot the body

    hover = HoverTool(tooltips=[
        ('date', '@open_time{%F %T}'),
        ('open', '@open{0.0000f}'),
        ('high', '@high{0.0000f}'),
        ('low', '@low{0.0000f}'),
        ('close', '@close{0.0000f}')
    ],
        formatters={'@open_time': 'datetime'})
    p.add_tools(hover)
2

Hi @BiggestSmile ,

Can you post a minimal, reproducible example (with sample data) so that someone can copy and paste your code and run it to see what you’re seeing?

3

I’ve managed to do something about it. I’ve now got correct dates at x axis. However, my “vbar” part is missing from the plot for some reason

4

Found out that vbars are there, just it’s scale is very small
Changed to 200000 to look like 1 with the default x-axis
I guess it’s due to being default ticker size of ms & 1 of width
Is there any way to account for this without the huge numbers?

5

Datetime axes are always denominated in ms since epoch. IIRC you can supply an actual datetime interval object for the width and it be converted automatically. But if you need/want to pass a plain number, it needs to be in the correct units (ms).