torch_timeseries.dataset.ETTm2

class ETTm2(root: str = '~/.torchtimeseries/data', columns=[])[source]

Bases: TimeSeriesDataset

Electricity Transformer Temperature minutely~(15 minutes level). This dataset contains 2-year data from two separate counties in China. Each data point consists of the target value “oil temperature” and 6 power load features.

This dataset is collected by Informer and is available at https://github.com/zhouhaoyi/ETDataset.

name

Name of the dataset.

Type:: str

freq

Frequency of the data points.

Type:: str

num_features

Number of features in the dataset.

Type:: int

length

Length of the dataset.

Type:: int

download()[source]: Downloads and extracts the dataset.

_load()[source]: Loads the dataset into a NumPy array.

_process(): Processes the dataset using the parent class’s process method.

Parameters:: root (str) – data save location, defaults to ~/.torchtimeseries/data