line gaps if time series interval is irregular

[miloaissatu]

python3.11, plotly_resampler 0.9.2

import numpy as np
import pandas as pd

from plotly_resampler import register_plotly_resampler

register_plotly_resampler(mode='widget')
pd.options.plotting.backend = 'plotly'


x = pd.DataFrame(np.random.normal(0, 0.01, 4000))
x.index = pd.Timestamp('2023-01-01') + (pd.Timedelta('1s') * np.random.rand(4000) * 20)
x = x.sort_index()
x[0] = x[0].cumsum()

x.plot(line_shape='hv')

when plotting a time series with irregular intervals there's line gaps when zooming is. is this expected behaviour? are there any configs to have it connect the datapoints?

[jonasvdd]

Hi @miloaissatu,

Yes there is!
You can use the NoGapHandler (docs), which can be passed to the default_gap_handler property of AbstractFigureAggregator its constructor (docs).

import numpy as np; import pandas as pd

from plotly_resampler import register_plotly_resampler
# NOTE: I added this import to the original example
from plotly_resampler.aggregation import NoGapHandler


# And added the `default_gap_handler` argument to the `register_plotly_resampler` call
register_plotly_resampler(mode='widget', default_gap_handler=NoGapHandler())
pd.options.plotting.backend = 'plotly'


x = pd.DataFrame(np.random.normal(0, 0.01, 4000))
x.index = pd.Timestamp('2023-01-01') + (pd.Timedelta('1s') * np.random.rand(4000) * 20)
x = x.sort_index()
x[0] = x[0].cumsum()

x.plot(line_shape='hv')

Hope this resolves your issue,
Jonas

[miloaissatu]

awesome. tyvm @jonasvdd

[boxingshark]

That helped me a lot! Thanks!