Time Series

This class represents a univariate time series. It allows for the storage of time indexed values as well as metadata. The constructor guarantees the monotonic increasing character of a time series.

Warning

The aim of a TimeSeries object is to be immutable

Constructor

TimeSeries(series, …)

A TimeSeries object is a series of time indexed values.

Methods

create(start, end[, freq, metadata])	Creates an empty TimeSeries object with the period as index
plot(*args, **kwargs)	Plot a TimeSeries
copy([deep])	Copy a TimeSeries
split_at(timestamp)	Split a TimeSeries at a defined point and include the splitting point in both as in [start,…,at] and [at,…,end].
split_in_chunks(n)	Split a TimeSeries into chunks of length n
fill(value)	Fill a TimeSeries with values
empty()	Empty the TimeSeries (fill all values with NaNs)
pad(limit[, side, value])	Pad a TimeSeries until a given limit
trim([side])	Remove NaNs from a TimeSeries start, end or both
merge(ts)	Merge two time series and make sure all the given indexes are sorted.

Processing

apply(func[, ts])	Wrapper around the Pandas apply function
resample(freq[, method])	Convert TimeSeries to specified frequency.
group_by(freq[, method])	Groups values by a frequency.
interpolate(*args, **kwargs)	Wrapper around the Pandas interpolate() method.
normalize(method)	Normalize a TimeSeries with a given method
round(decimals)	Round the values in the series.values
sort(*args, **kwargs)	Sort a TimeSeries by time stamps

Analysis

min()	Get the minimum value of a TimeSeries
max()	Get the maximum value of a TimeSeries
mean()	Get the mean value of a TimeSeries
median()	Get the median value of a TimeSeries
skewness()	Get the skewness of a TimeSeries
kurtosis()	Get the kurtosis of a TimeSeries
describe([percentiles, include, exclude])	Describe a TimeSeries with the describe function from Pandas
start()	Get the first Timestamp of a TimeSeries
end()	Get the last Timestamp of a TimeSeries
boundaries()	Get a tuple with the TimeSeries first and last index
frequency()	Get the frequency of a TimeSeries
duration()	Get the duration of the TimeSeries

I/O

to_text(path)	Export a TimeSeries to text format
to_array()	Convert a TimeSeries to Numpy Array
to_pickle(path)	Export a TimeSeries to Pickle
to_darts()	Convert a TimeSeries to Darts TimeSeries
to_df()	Converts a TimeSeries to a Pandas DataFrame

timeatlas.read_write.read_text timeatlas.time_series.TimeSeries