Create a temporal stream from a data source.
TemporalDataStream takes the whole data set containing the X (features),
time (timestamps) and Y (targets).
The features and targets or only the features if they are passed
in the y parameter.
The timestamp column of each instance. If its a pandas.Series, it will
be converted into a numpy.ndarray. If None, delay by number of samples
is considered and sample_delay must be int.
The column index from which the targets start.
The number of targets.
A list of indices corresponding to the location of categorical features.
A string to id the data.
If True, consider that data, y, and time are already ordered by timestamp.
Otherwise, the data is ordered based on time timestamps (time cannot be
If True, allows NaN values in the data. Otherwise, an error is raised.
The stream object provides upon request a number of samples, in a way such
that old samples cannot be accessed at a later time. This is done to
correctly simulate the stream context.
Retrieves minimum information from the stream
Collects and returns the information about the configuration of the estimator
Get parameters for this estimator.
Checks if stream has more samples.
Determine if the stream is restartable.
Retrieves last batch_size samples in the stream.
Returns the estimated number of remaining samples.
Get next sample.
Prepare the stream for use.
Prints all the samples in the stream.
Resets the estimator to its initial state.
Restarts the stream.
Set the parameters of this estimator.
Return the features’ columns.
Get the list of the categorical features index.
Return the data set used to generate the stream.
Retrieve the names of the features.
Retrieve the number of integer features.
Retrieve the number of features.
Retrieve the number of numerical features.
Get the number of targets.
Get the number of the column where Y begins.
Retrieve the names of the targets
Retrieve all target_values in the stream for each target.
Return the targets’ columns.
the features’ columns
List of categorical features index.
names of the features
Used by evaluator methods to id the stream.
The default format is: ‘Stream name - n_targets, n_classes, n_features’.
Stream data information
Configuration of the estimator.
If True, will return the parameters for this estimator and
contained subobjects that are estimators.
Parameter names mapped to their values.
True if stream has more samples.
True if stream is restartable.
A numpy.ndarray of shape (batch_size, n_features) and an array-like of shape
(batch_size, n_targets), representing the next batch_size samples.
The number of integer features in the stream.
The total number of features.
The number of numerical features in the stream.
Remaining number of samples.
Get the number of targets.
If there is enough instances to supply at least batch_size samples,
those are returned. If there aren’t a tuple of (None, None) is returned.
The number of instances to return.
Returns the next batch_size instances (sample_x, sample_y,
sample_time, sample_delay (if available), sample_weight
(if available)). For general purposes the return can be
treated as a numpy.ndarray.
Deprecated in v0.5.0 and will be removed in v0.7.0
It basically server the purpose of reinitializing the stream to
its initial state.
The method works on simple estimators as well as on nested objects
(such as pipelines). The latter have parameters of the form
<component>__<parameter> so that it’s possible to update each
component of a nested object.
The number of the column where Y begins.
the names of the targets in the stream.
list of lists of all target_values for each target
the targets’ columns