ts_datasets.forecast package

Datasets for time series forecasting. Really, these are just time series with no labels of any sort.

ts_datasets.forecast.get_dataset(dataset_name, rootdir=None)
Parameters

  • dataset_name (str) – the name of the dataset to load, formatted as <name> or <name>_<subset>, e.g. EnergyPower or M4_Hourly

  • rootdir (Optional[str]) – the directory where the desired dataset is stored. Not required if the package ts_datasets is installed in editable mode, i.e. with flag -e.

Return type

BaseDataset

Returns

the data loader for the desired dataset (and subset) desired

class ts_datasets.forecast.M4(subset='Hourly', rootdir=None)

Bases: BaseDataset

The M4 Competition data is an extended and diverse set of time series to identify the most accurate forecasting method(s) for different types of domains, including Business, financial and economic forecasting, and different type of granularity, including Yearly (23,000 sequences), Quarterly (24,000 sequences), Monthly (48,000 sequences), Weekly(359 sequences), Daily (4,227 sequences) and Hourly (414 sequences) data.

valid_subsets = ['Yearly', 'Quarterly', 'Monthly', 'Weekly', 'Daily', 'Hourly']
url = 'https://github.com/Mcompetitions/M4-methods/raw/master/Dataset/{}.csv'
time_series: list

A list of all individual time series contained in the dataset. Iterating over the dataset will iterate over this list. Note that for some large datasets, time_series may be a list of filenames, which are read lazily either during iteration, or whenever __getitem__ is invoked.

metadata: list

A list containing the metadata for all individual time series in the dataset.

class ts_datasets.forecast.EnergyPower(rootdir=None)

Bases: BaseDataset

Wrapper to load the open source energy grid power usage dataset.

Parameters

rootdir – The root directory at which the dataset can be found.

time_series: list

A list of all individual time series contained in the dataset. Iterating over the dataset will iterate over this list. Note that for some large datasets, time_series may be a list of filenames, which are read lazily either during iteration, or whenever __getitem__ is invoked.

metadata: list

A list containing the metadata for all individual time series in the dataset.

class ts_datasets.forecast.SeattleTrail(rootdir=None)

Bases: BaseDataset

Wrapper to load the open source Seattle Trail pedestrian/bike traffic dataset.

Parameters

rootdir – The root directory at which the dataset can be found.

time_series: list

A list of all individual time series contained in the dataset. Iterating over the dataset will iterate over this list. Note that for some large datasets, time_series may be a list of filenames, which are read lazily either during iteration, or whenever __getitem__ is invoked.

metadata: list

A list containing the metadata for all individual time series in the dataset.

class ts_datasets.forecast.SolarPlant(rootdir=None, num_columns=100)

Bases: BaseDataset

Wrapper to load the open source solar plant power dataset.

Note

The loader currently only includes the first 100 (of 405) variables.

Parameters

  • rootdir – The root directory at which the dataset can be found.

  • num_columns – indicates how many univariate columns should be returned

time_series: list

A list of all individual time series contained in the dataset. Iterating over the dataset will iterate over this list. Note that for some large datasets, time_series may be a list of filenames, which are read lazily either during iteration, or whenever __getitem__ is invoked.

metadata: list

A list containing the metadata for all individual time series in the dataset.