ramjet.logging.wandb_logger¶

Classes¶

class ExampleRequest[source]¶

A representation of a request for an example.

class WandbLoggable[source]¶

Bases: abc.ABC

An object which is loggable to wandb.

log(self, summary_name: str, epoch: int)[source]¶

Logs self to wandb.

Parameters:	summary_name – The name of the summary to use on wandb. epoch – The current epoch to log with.

static log_figure(summary_name: str, figure: plotly.graph_objects.Figure, epoch: int)[source]¶

Logs a figure to wandb.

Parameters:	summary_name – The name of the summary to use on wandb. figure – The figure to be logged. epoch – The current epoch to log with.

class WandbLoggableLightCurve(light_curve_name: str, light_curve: LightCurve)[source]¶

Bases: ramjet.logging.wandb_logger.WandbLoggable

A wandb loggable light curve.

__init__(self, light_curve_name: str, light_curve: LightCurve)[source]¶

Initialize self. See help(type(self)) for accurate signature.

log(self, summary_name: str, epoch: int)[source]¶

Logs self to wandb.

Parameters:	summary_name – The name of the summary to use on wandb. epoch – The current epoch to log with.

class WandbLoggableInjection[source]¶

Bases: ramjet.logging.wandb_logger.WandbLoggable

A wandb loggable containing logging data for injecting a signal into a light curve.

__init__(self)[source]¶

Initialize self. See help(type(self)) for accurate signature.

log(self, summary_name: str, epoch: int)[source]¶

Logs self to wandb.

Parameters:	summary_name – The name of the summary to use on wandb. epoch – The current epoch to log with.

class WandbLogger[source]¶

A class to log to wandb.

loggable_types¶

__init__(self)[source]¶

Initialize self. See help(type(self)) for accurate signature.

classmethod new(cls, entity: Optional[str] = None, project: Optional[str] = None)[source]¶

Creates a new logger.

Returns:	The logger.

process_py_mapper_example_queues(self, epoch: int)[source]¶

Processes the example queues, logging the items to wandb.

request_examples_from_py_mapper_processes(self)[source]¶

Sends requests for examples to the other processes.

create_callback(self)[source]¶

Creates a callback for the fit loop to call the logger methods.

Returns:	The callback.

create_request_queue_for_collection(self, name: str)[source]¶

Creates a queue to send requests for examples on.

Parameters:	name – The name of the queue.
Returns:	The queue.

create_example_queue_for_collection(self, name: str)[source]¶

Creates a queue to receive examples on.

Parameters:	name – The name of the queue.
Returns:	The queue.

static should_produce_example(request_queue: multiprocess.Queue)[source]¶

Checks a request queue to see if an example has been requested.

Parameters:	request_queue – The request queue.
Returns:	Whether an example was requested on the queue.

static submit_loggable(example_queue: multiprocess.Queue, loggable: WandbLoggable)[source]¶

Submits a loggable to a request queue to be logged by the main process.

Parameters:	example_queue – The queue to submit the loggable to. loggable – The loggable to be submitted.

class WandbLoggerCallback(logger: WandbLogger)[source]¶

Bases: tensorflow.keras.callbacks.Callback

A callback for the training loop to call to utilize the wandb logger.

__init__(self, logger: WandbLogger)[source]¶

on_epoch_begin(self, epoch, logs=None)[source]¶

Called at the beginning of an epoch.

on_epoch_end(self, epoch, logs=None)[source]¶

Called at the end of an epoch.

static is_power(number: int, base: int)[source]¶

Checks if number is a power of the given base.

Parameters:	number – The number. base – The base.
Returns:	Whether the number is a power of the base.