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, **kwargs)
- Parameters
dataset_name (
str
) – the name of the dataset to load, formatted as<name>
or<name>_<subset>
, e.g.EnergyPower
orM4_Hourly
rootdir (
Optional
[str
]) – the directory where the desired dataset is stored. Not required if the packagets_datasets
is installed in editable mode, i.e. with flag-e
.kwargs – keyword arguments for the data loader you are trying to load.
- Return type
- Returns
the data loader for the desired dataset (and subset) desired
- class ts_datasets.forecast.CustomDataset(rootdir, test_frac=0.5, time_col=None, time_unit='s', data_cols=None, index_cols=None)
Bases:
BaseDataset
Wrapper to load a custom dataset. Please review the tutorial to get started.
- Parameters
rootdir – Filename of a single CSV, or a directory containing many CSVs. Each CSV must contain 1 or more time series.
test_frac – If we don’t find a column “trainval” in the time series, this is the fraction of each time series which we use for testing.
time_col – Name of the column used to index time. We use the first non-index, non-metadata column if none is given.
time_unit – If the time column is numerical, we assume it is a timestamp expressed in this unit.
data_cols – Name of the columns to fetch from the dataset. If
None
, use all non-time, non-index columns.index_cols – If a CSV file contains multiple time series, these are the columns used to index those time series. For example, a CSV file may contain time series of sales for many (store, department) pairs. In this case,
index_cols
may be["Store", "Dept"]
. The values of the index columns will be added to the metadata of the data loader.
- property metadata_cols
- check_ts_for_metadata(ts, col)
- 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.
source: https://github.com/Mcompetitions/M4-methods/tree/master/Dataset
timeseries sequences: 100,000
- valid_subsets = ['Yearly', 'Quarterly', 'Monthly', 'Weekly', 'Daily', 'Hourly']
- url = 'https://github.com/Mcompetitions/M4-methods/raw/master/Dataset/{}.csv'
- class ts_datasets.forecast.EnergyPower(rootdir=None)
Bases:
BaseDataset
Wrapper to load the open source energy grid power usage dataset.
source: https://www.kaggle.com/robikscube/hourly-energy-consumption
contains one 10-variable time series
- Parameters
rootdir – The root directory at which the dataset can be found.
- class ts_datasets.forecast.SeattleTrail(rootdir=None)
Bases:
BaseDataset
Wrapper to load the open source Seattle Trail pedestrian/bike traffic dataset.
source: https://www.kaggle.com/city-of-seattle/seattle-burke-gilman-trail
contains one 5-variable time series
- Parameters
rootdir – The root directory at which the dataset can be found.
- class ts_datasets.forecast.SolarPlant(rootdir=None, num_columns=100)
Bases:
BaseDataset
Wrapper to load the open source solar plant power dataset.
contains one 405-variable time series
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