Plotting with log scale

[andreas-vester]

How can I turn the scale of the y-axis into a log-scale? In plotly, you can make use of the log_y=True keyword argument. I am expecting to be able to use it with vectorbt as well. I probably need a little push into the right direction.

df["close"].vbt.plot(trace_kwargs=dict(name="Closing price"))

[polakowo]

import vectorbt as vbt
import pandas as pd

pd.Series([1, 2, 3]).vbt.plot(yaxis=dict(type="log"))

[andreas-vester]

Where am I supposed to find this information in the docs?

[polakowo]

First of all, vbt attribute on top of a Series/DataFrame is an accessor. You can view this page to see what accessors are supported and how they are accessed. Since we called plot directly on vbt, we search for a method called plot of the most generic accessor (GenericAccessor): here. As per docs, we see that whatever arguments we pass to this method, they are forwarded down to Scatter. Since xaxis(type="log") is a layout property, we pass it as part of **layout_kwargs, that is, directly. If any argument is no listed in the documentation or description is rather vague, you can always view the source (there is a button under each method) to get an idea of what's going on. I know, the documentation is not sexy, but it's the best I could organize for such a large and complex codebase.

[andreas-vester]

All right, I had been almost there. All right, I had been almost there. I thought that one has to put it as a kwarg, so I browsed the plotly docs and came across log_y=True (https://plotly.com/python/log-plot/).

Anyway, I highly appreciate the effort you put into vectorbt's docs. However, it is still somewhat challenging to get an overview of the vast functionality when you are new to this package. So getting support from you is invaluable!! - Thanks!