mednet.engine.loggers¶
Custom lightning loggers.
Classes
|
Custom implementation implementation of lightning's TensorboardLogger. |
- class mednet.engine.loggers.CustomTensorboardLogger(save_dir, name='lightning-logs', version=None, log_graph=False, default_hp_metric=True, prefix='', sub_dir=None, **kwargs)[source]¶
Bases:
TensorBoardLogger
Custom implementation implementation of lightning’s TensorboardLogger.
This implementation puts all logs inside the same directory, instead of a separate “version_n” directories, which is the default lightning behaviour.
- Parameters:
save_dir (
Union
[str
,Path
]) – Directory where to save the logs to.name (
str
) – Experiment name. Defaults todefault
. If it is the empty string then no per-experiment subdirectory is used.version (
int
|str
|None
) – Experiment version. If version is not specified the logger inspects the save directory for existing versions, then automatically assigns the next available version. If it is a string then it is used as the run-specific subdirectory name, otherwiseversion_${version}
is used.log_graph (
bool
) – Adds the computational graph to tensorboard. This requires that the user has defined the self.example_input_array attribute in their model.default_hp_metric (
bool
) – Enables a placeholder metric with key hp_metric when log_hyperparams is called without a metric (otherwise calls to log_hyperparams without a metric are ignored).prefix (
str
) – A string to put at the beginning of metric keys.sub_dir (
Union
[str
,Path
,None
]) – Sub-directory to group TensorBoard logs. If a sub_dir argument is passed then logs are saved in/save_dir/name/version/sub_dir/
. Defaults toNone
in which logs are saved in/save_dir/name/version/
.**kwargs (
dict
[str
,Any
]) – Additional arguments used bytensorboardX.SummaryWriter
can be passed as keyword arguments in this logger. To automatically flush to disk,max_queue
sets the size of the queue for pending logs before flushing.flush_secs
determines how many seconds elapses before flushing.