keras_ Callback function summary

keras The callback function summarizes various problems that may occur in the actual project process , And made targeted measures , Very useful , So it was recorded , In the future, if there is a new official API, I will update it regularly .

class BaseLogger: Callback function of cumulative evaluation value

Callback that accumulates epoch averages of metrics.

class CSVLogger: Save the flow data of each iteration to csv File callback function

Callback that streams epoch results to a csv file.

class Callback: Abstract base class for generating new callbacks

Abstract base class used to build new callbacks.

class EarlyStopping: Stop training when the monitoring quantity stops improving

Stop training when a monitored quantity has stopped improving.

class History: Log to History Callback function for

Callback that records events into a History object.

class LambdaCallback: Callback function for dynamically creating simple custom callbacks

Callback for creating simple, custom callbacks on-the-fly.

class LearningRateScheduler: Learning rate scheduler

Learning rate scheduler.

class ModelCheckpoint: At every epoch Then save the model

Save the model after every epoch.

class ProgbarLogger: Print evaluation values to stdout Callback function for

Callback that prints metrics to stdout.

class ReduceLROnPlateau: Reduce learning when an assessment stops improving

Reduce learning rate when a metric has stopped improving.

class RemoteMonitor: Callback used to stream events to the server

Callback used to stream events to a server.

class TensorBoard: Enable TensorBoard Visualization

Enable visualizations for TensorBoard.

class TerminateOnNaN: When you meet NaN The callback to terminate the training in case of loss

Callback that terminates training when a NaN loss is encountered.