tf.keras.preprocessing.sequence.TimeseriesGenerator

Utility class for generating batches of temporal data.

Inherits From: PyDataset

DEPRECATED.

This class takes in a sequence of data-points gathered at equal intervals, along with time series parameters such as stride, length of history, etc., to produce batches for training/validation.

dataIndexable generator (such as list or Numpy array) containing consecutive data points (timesteps). The data should be at 2D, and axis 0 is expected to be the time dimension.
targetsTargets corresponding to timesteps in data. It should have same length as data.
lengthLength of the output sequences (in number of timesteps).
sampling_ratePeriod between successive individual timesteps within sequences. For rate r, timesteps data[i], data[i-r], ... data[i - length] are used for create a sample sequence.
stridePeriod between successive output sequences. For stride s, consecutive output samples would be centered around data[i], data[i+s], data[i+2*s], etc.
start_indexData points earlier than start_index will not be used in the output sequences. This is useful to reserve part of the data for test or validation.
end_indexData points later than end_index will not be used in the output sequences. This is useful to reserve part of the data for test or validation.
shuffleWhether to shuffle output samples, or instead draw them in chronological order.
reverseBoolean: if true, timesteps in each output sample will be in reverse chronological order.
batch_sizeNumber of timeseries samples in each batch (except maybe the last one).

A PyDataset instance.

max_queue_size

num_batchesNumber of batches in the PyDataset.
use_multiprocessing

workers

Methods

get_config

View source

Returns the TimeseriesGenerator configuration as Python dictionary.

Returns
A Python dictionary with the TimeseriesGenerator configuration.

on_epoch_end

View source

Method called at the end of every epoch.

to_json

View source

Returns a JSON string containing the generator's configuration.

Args
**kwargsAdditional keyword arguments to be passed to json.dumps().

Returns
A JSON string containing the tokenizer configuration.

__getitem__

View source

Gets batch at position index.

Args
indexposition of the batch in the PyDataset.

Returns
A batch

__len__

View source