write_dataframe_to_tsfile

write_dataframe_to_tsfile(data, path, problem_name='sample_data', class_label=None, class_value_list=None, equal_length=False, series_length=-1, missing_values='NaN', comment=None, fold='')

Output a dataset in dataframe format to .ts file.

Parameters:

data: pandas dataframe

The dataset in a dataframe to be written as a ts file which must be of the structure specified in the documentation examples/loading_data.ipynb. index | dim_0 | dim_1 | … | dim_c-1

0 | pd.Series | pd.Series | pd.Series | pd.Series 1 | pd.Series | pd.Series | pd.Series | pd.Series

… | … | … | … | …

n | pd.Series | pd.Series | pd.Series | pd.Series

path: str

The full path to output the ts file to.

problem_name: str, default=”sample_data”

The problemName to print in the header of the ts file and also the name of the file.

class_label: list of str or None, default=None

The problems class labels to show the possible class values for in the file header, optional.

class_value_list: list, ndarray or None, default=None

The class values for each case, optional.

equal_length: bool, default=False

Indicates whether each series is of equal length.

series_length: int, default=-1

Indicates the series length if they are of equal length.

missing_values: str, default=”NaN”

Representation for missing values.

comment: str or None, default=None

Comment text to be inserted before the header in a block.

fold: str or None, default=None

Addon at the end of the filename, i.e. _TRAIN or _TEST.

Returns:

None

Notes

This version currently does not support writing timestamp data.