tf.keras.preprocessing.timeseries_dataset_from_array

Creates a dataset of sliding windows over a timeseries provided as array.

Main aliases

tf.keras.utils.timeseries_dataset_from_array

Used in the notebooks

Used in the tutorials

This function takes in a sequence of data-points gathered at equal intervals, along with time series parameters such as length of the sequences/windows, spacing between two sequence/windows, etc., to produce batches of timeseries inputs and targets.

data Numpy array or eager tensor containing consecutive data points (timesteps). Axis 0 is expected to be the time dimension.
targets Targets corresponding to timesteps in data. targets[i] should be the target corresponding to the window that starts at index i (see example 2 below). Pass None if you don't have target data (in this case the dataset will only yield the input data).
sequence_length Length of the output sequences (in number of timesteps).
sequence_stride Period between successive output sequences. For stride s, output samples would start at index data[i], data[i + s], data[i + 2 * s], etc.
sampling_rate Period between successive individual timesteps within sequences. For rate r, timesteps data[i], data[i + r], ... data[i + sequence_length] are used for creating a sample sequence.
batch_size Number of timeseries samples in each batch (except maybe the last one). If None, the data will not be batched (the dataset will yield individual samples).
shuffle Whether to shuffle output samples, or instead draw them in chronological order.
seed Optional int; random seed for shuffling.
start_index Optional int; data points earlier (exclusive) 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_index Optional int; data points later (exclusive) than end_index will not be used in the output sequences. This is useful to reserve part of the data for test or validation.

A tf.data.Dataset instance. If targets was passed, the dataset yields tuple (batch_of_sequences, batch_of_targets). If not, the dataset yields only batch_of_sequences.

Example 1:

Consider indices [0, 1, ... 98]. With sequence_length=10, sampling_rate=2, sequence_stride=3, shuffle=False, the dataset will yield batches of sequences composed of the following indices:

First sequence:  [0  2  4  6  8 10 12 14 16 18]
Second sequence: [3  5  7  9 11 13 15 17 19 21]
Third sequence:  [6  8 10 12 14 16 18 20 22 24]
...
Last sequence:   [78 80 82 84 86 88 90 92 94 96]

In this case the last 2 data points are discarded since no full sequence can be generated to include them (the next sequence would have started at index 81, and thus its last step would have gone over 98).

Example 2: Temporal regression.

Consider an array data of scalar values, of shape (steps,). To generate a dataset that uses the past 10 timesteps to predict the next timestep, you would use:

input_data = data[:-10]
targets
= data[10:]
dataset
= timeseries_dataset_from_array(
    input_data
, targets, sequence_length=10)
for batch in dataset:
  inputs
, targets = batch
 
assert np.array_equal(inputs[0], data[:10])  # First sequence: steps [0-9]
 
# Corresponding target: step 10
 
assert np.array_equal(targets[0], data[10])
 
break

Example 3: Temporal regression for many-to-many architectures.

Consider two arrays of scalar values X and Y, both of shape (100,). The resulting dataset should consist samples with 20 timestamps each. The samples should not overlap. To generate a dataset that uses the current timestamp to predict the corresponding target timestep, you would use:

X = np.arange(100)
Y
= X*2

sample_length
= 20
input_dataset
= timeseries_dataset_from_array(
    X
, None, sequence_length=sample_length, sequence_stride=sample_length)
target_dataset
= timeseries_dataset_from_array(
    Y
, None, sequence_length=sample_length, sequence_stride=sample_length)

for batch in zip(input_dataset, target_dataset):
    inputs
, targets = batch
   
assert np.array_equal(inputs[0], X[:sample_length])

   
# second sample equals output timestamps 20-40
   
assert np.array_equal(targets[1], Y[sample_length:2*sample_length])
   
break