physioex.train.logger.Logger#
- class physioex.train.logger.Logger[source]#
Bases:
ABCBackend-agnostic experiment logger.
Every method has a no-op default so backends only override what they support;
log()is the single required hook.- __init__()#
Methods
__init__()close()Release backend resources at the end of training.
log(stage, epoch, step, loss[, accuracy, ...])Log the scalar metrics of a single train/validation step.
log_figure(tag, figure, step)Record a matplotlib figure (e.g. a confusion matrix).
log_graph(model, input_sample)Record the model computational graph.
log_histogram(tag, values, step)Record a distribution of values (e.g. weights or gradients).
log_hparams(hparams[, metrics])Record run hyper-parameters once at the start of training.
log_learning_rate(step, value)Record the learning rate used at a global step.
log_scalars(mapping, step)Record a batch of named scalars (per-class metrics, aggregates).
log_update_norm(step, value)Record the L2 norm of the optimizer update at a global step.
update()Flush buffered events to the backend.
watch_model(model[, log_freq])Automatically track parameters/gradients (W&B only).
- abstract log(stage, epoch, step, loss, accuracy=None, extra_metrics=None)[source]#
Log the scalar metrics of a single train/validation step.
- log_histogram(tag, values, step)[source]#
Record a distribution of values (e.g. weights or gradients).
- log_hparams(hparams, metrics=None)[source]#
Record run hyper-parameters once at the start of training.
- log_scalars(mapping, step)[source]#
Record a batch of named scalars (per-class metrics, aggregates).