tslearn.utils.to_pyflux_dataset

tslearn.utils.to_pyflux_dataset(X)[source]

Transform a tslearn-compatible dataset into a pyflux dataset.

Parameters:
X: array, shape = (n_ts, sz, d), where n_ts=1

tslearn-formatted dataset to be cast to pyflux format

Returns:
Pandas data-frame

pyflux-formatted dataset (cf. link)

Notes

Conversion from/to pyflux format requires pandas to be installed.

Examples

>>> tslearn_arr = numpy.random.randn(1, 16, 1)
>>> pyflux_df = to_pyflux_dataset(tslearn_arr)
>>> pyflux_df.shape
(16, 1)
>>> pyflux_df.columns[0]
'dim_0'
>>> tslearn_arr = numpy.random.randn(1, 16, 2)
>>> pyflux_df = to_pyflux_dataset(tslearn_arr)
>>> pyflux_df.shape
(16, 2)
>>> pyflux_df.columns[1]
'dim_1'
>>> tslearn_arr = numpy.random.randn(10, 16, 1)
>>> to_pyflux_dataset(tslearn_arr)  # doctest: +IGNORE_EXCEPTION_DETAIL
Traceback (most recent call last):
...
ValueError: Array should be made of a single time series (10 here)