noggin.logger.LiveLogger¶

class noggin.logger.LiveLogger(*args, **kwargs)[source]¶

Logs batch-level and epoch-level summary statistics of the training and testing metrics of a model during a session.

Examples

A simple example in which we log two iterations of training batches, and set an epoch.

>>> from noggin import LiveLogger
>>> logger = LiveLogger()
>>> logger.set_train_batch(dict(metric_a=2., metric_b=1.), batch_size=10)
>>> logger.set_train_batch(dict(metric_a=0., metric_b=2.), batch_size=4)
>>> logger.set_train_epoch()  # compute the mean statistics
>>> logger
LiveLogger(metric_a, metric_b)
number of training batches set: 2
number of training epochs set: 1
number of testing batches set: 0
number of testing epochs set: 0

Accessing our logged batch-level and epoch-level data

>>> logger.to_xarray("train")
MetricArrays(batch=<xarray.Dataset>
Dimensions:     (iterations: 2)
Coordinates:
  * iterations  (iterations) int32 1 2
Data variables:
    metric_a    (iterations) float64 2.0 0.0
    metric_b    (iterations) float64 1.0 2.0,
epoch=<xarray.Dataset>
Dimensions:     (iterations: 1)
Coordinates:
  * iterations  (iterations) int32 2
Data variables:
    metric_a    (iterations) float64 1.429
    metric_b    (iterations) float64 1.286)

Attributes:	`test_metrics` The batch and epoch data for each test-metric. `train_metrics` The batch and epoch data for each train-metric.

Methods

`from_dict`(logger_dict, Any])	Records the state of the logger in a dictionary.
`set_test_batch`(metrics, numbers.Real], …)	Record batch-level measurements for test-metrics.
`set_test_epoch`()	Record an epoch for the test-metrics.
`set_train_batch`(metrics, numbers.Real], …)	Record batch-level measurements for train-metrics.
`set_train_epoch`()	Record an epoch for the train-metrics.
`to_dict`()	Records the state of the logger in a dictionary.
`to_xarray`(train_or_test)	Returns xarray datasets for the batch-level and epoch-level metrics, respectively, for either the train-metrics or test-metrics.

__init__(*args, **kwargs)[source]¶: LiveLogger.__init__ does not utilize any input arguments, but accepts *args, **kwargs so that it can be used as a drop-in replacement for

LivePlot.

Methods

`__init__`(args, *kwargs)	LiveLogger.__init__ does not utilize any input arguments, but accepts `args, *kwargs` so that it can be used as a drop-in replacement for `LivePlot`.
`from_dict`(logger_dict, Any])	Records the state of the logger in a dictionary.
`set_test_batch`(metrics, numbers.Real], …)	Record batch-level measurements for test-metrics.
`set_test_epoch`()	Record an epoch for the test-metrics.
`set_train_batch`(metrics, numbers.Real], …)	Record batch-level measurements for train-metrics.
`set_train_epoch`()	Record an epoch for the train-metrics.
`to_dict`()	Records the state of the logger in a dictionary.
`to_xarray`(train_or_test)	Returns xarray datasets for the batch-level and epoch-level metrics, respectively, for either the train-metrics or test-metrics.

Attributes

`test_metrics`	The batch and epoch data for each test-metric.
`train_metrics`	The batch and epoch data for each train-metric.