edflow.hooks.logging_hooks.tf_logging_hook module¶
Reference¶
-
class
edflow.hooks.logging_hooks.tf_logging_hook.LoggingHook(scalars={}, histograms={}, images={}, logs={}, graph=None, interval=100, root_path='logs', log_images_to_tensorboard=False)[source]¶ Bases:
edflow.hooks.hook.HookSupply and evaluate logging ops at an intervall of training steps.
-
__init__(scalars={}, histograms={}, images={}, logs={}, graph=None, interval=100, root_path='logs', log_images_to_tensorboard=False)[source]¶ - Parameters
scalars (dict) – Scalar ops.
histograms (dict) – Histogram ops.
images (dict) – Image ops. Note that for these no tensorboard logging ist used but a custom image saver.
logs (dict) – Logs to std out via logger.
graph (tf.Graph) – Current graph.
interval (int) – Intervall of training steps before logging.
root_path (str) – Path at which the logs are stored.
-
before_epoch(ep)[source]¶ Called before each epoch.
- Parameters
epoch (int) – Index of epoch that just started.
-
before_step(batch_index, fetches, feeds, batch)[source]¶ Called before each step. Can update any feeds and fetches.
- Parameters
step (int) – Current training step.
fetches (list or dict) – Fetches for the next session.run call.
feeds (dict) – Data used at this step.
batch (list or dict) – All data available at this step.
-
-
class
edflow.hooks.logging_hooks.tf_logging_hook.ImageOverviewHook(images={}, interval=100, root_path='logs')[source]¶ Bases:
edflow.hooks.hook.Hook-
__init__(images={}, interval=100, root_path='logs')[source]¶ Logs an overview of all image outputs at an intervall of training steps.
- Parameters
scalars (dict) – Scalar ops.
histograms (dict) – Histogram ops.
images (dict) – Image ops. Note that for these no tensorboard logging ist used but a custom image saver.
logs (dict) – Logs to std out via logger.
graph (tf.Graph) – Current graph.
interval (int) – Intervall of training steps before logging.
root_path (str) – Path at which the logs are stored.
-