Module: tf.keras.callbacks
class BaseLogger: 累計評估值的回調函數
Callback that accumulates epoch averages of metrics. class CSVLogger: 將每次迭代的流數據保存到csv文件的回調函數
Callback that streams epoch results to a csv file. class Callback: 用於生成新回調的抽象基類
Abstract base class used to build new callbacks. class EarlyStopping: 當監控量停止改善時停止培訓
Stop training when a monitored quantity has stopped improving. class History: 記錄日志到History的回調函數
Callback that records events into a History object. class LambdaCallback: 用於動態創建簡單的自定義回調的回調函數
Callback for creating simple, custom callbacks on-the-fly. class LearningRateScheduler: 學習率調度程序
Learning rate scheduler. class ModelCheckpoint: 在每個epoch之后保存模型
Save the model after every epoch. class ProgbarLogger: 將評估值打印到stdout的回調函數
Callback that prints metrics to stdout. class ReduceLROnPlateau: 當一個評估停止改進時降低學習率
Reduce learning rate when a metric has stopped improving. class RemoteMonitor: 用於將事件流式傳輸到服務器的回調
Callback used to stream events to a server. class TensorBoard: 啟用TensorBoard的可視化
Enable visualizations for TensorBoard. class TerminateOnNaN: 當遇到NaN丟失時終止培訓的回調
Callback that terminates training when a NaN loss is encountered.
class BaseLogger
該類的結構：
