I am using pandas v 1.5.3 and Python 3.10.
I have timeseries data along with flag columns like this:
Open High Low Close Adj Close Volume Flag
Date
2018-01-10 155.745697 157.103256 155.353729 156.959854 122.008064 4366109 1
2018-01-12 156.806885 157.495224 155.860428 155.965576 121.235191 5263367 3
2018-01-22 154.407272 156.768646 154.024857 155.449326 120.833931 8870917 1
2018-02-06 143.680695 148.652008 142.552582 148.508606 115.438744 10321614 8
2018-02-09 142.065002 143.919693 138.049713 142.934998 112.200180 8188402 1
Here are the data types of the columns in the dataframe.
Open float64
High float64
Low float64
Close float64
Adj Close float64
Volume int64
Flag int64
This is the index type:
Index dtype
datetime64[ns]
I want to use the flag number to determine the type of symbol to plot on the chart (a simple mapping from my boolean flag to a matplotlib marker symbol to use on the chart).
This is what I have so far (only relevant part shown):
# Define the plot style
ohlc_style = mpf.make_mpf_style(base_mpf_style='charles',
y_on_right=False,
marketcolors=mpf.make_marketcolors(up='g', down='r'),
mavcolors=['purple', 'orange'],
figcolor='w')
# Plot the chart with the 'Flag' as scatter points
mpf.plot(df, type='candle', style=ohlc_style, volume=True, show_nontrading=False, addplot=df['Flag'], scatter=True)
However, when I run this, it raises the following exception:
raise TypeError('kwarg "'+key+'" validator returned False for value: "'+str(value)+'"\n '+v)
TypeError: kwarg "addplot" validator returned False for value: "Date
2018-01-02 0
2018-01-03 0
2018-01-04 0
2018-01-05 0
2018-01-08 0
..
2021-12-27 0
2021-12-28 0
2021-12-29 0
2021-12-30 0
2021-12-31 0
Name: Flag, Length: 1008, dtype: int64"
'Validator' : lambda value: isinstance(value,dict) or (isinstance(value,list) and all([isinstance(d,dict) for d in value])) },
How do I fix this error, and also to use a mapping function that uses a hardocded map to map from my flag integers to a matplotlib.markers style?
As mentioned by @Corralien you cannot pass data directly to the addplot= kwarg, but rather you must call mpf.make_addplot() to generate and "addplot object" to pass as the value of the addplot= kwarg.
In addition to this, to have multiple scatter markers, do the following:
When calling mpf.make_addplot() set kwarg marker= to a list (or tuple) of markers (instead of a single marker value).
For some examples, click here.
Note that if marker= is not a single marker, then it must contain the same number of markers that the data frame contains rows.
Note also (as you can see in the examples) kwarg color= may also be either a single color, or a list or tuple of colors: If a list or tuple, then again it must contain the same number of elements as the data frame contains rows.