torch_timeseries.dataset.Weather

class Weather(root: str = './data')[source]

Bases: TimeSeriesDataset

This dataset contains local climatological data for nearly 1,600 in US locations, covering 4 years from 2010 to 2013. Data points are collected every hour. Each data point consists of the target value “wet bulb” and 11 climate features. The raw data can be acquired at https://www.ncei.noaa.gov/data/local-climatological-data/.

name

Name of the dataset.

Type:: str

num_features

Number of features in the dataset.

Type:: int

length

Length of the dataset.

Type:: int

freq

Frequency of the data points.

Type:: str

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

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

_summary_

Parameters:: root (str) – data save location