deephyper.contrib.callbacks

deephyper.contrib.callbacks

Functions

import_callback

Import a callback class from its name.

Classes

CSVExtendedLogger

Callback that streams epoch results to a csv file. Supports all values that can be represented as a string, including 1D iterables such as np.ndarray. Example: `python csv_logger = CSVLogger('training.log') model.fit(X_train, Y_train, callbacks=[csv_logger]) ` :param filename: filename of the csv file, e.g. ‘run/log.csv’. :param separator: string used to separate elements in the csv file. :param append: True: append if file exists (useful for continuing training). False: overwrite existing file,.

LearningRateScheduleCallback

LearningRateWarmupCallback

StopIfUnfeasible

TimeStopping

Stop training when a specified amount of time has passed. :param seconds: maximum amount of time before stopping. Defaults to 86400 (1 day). :param verbose: verbosity mode. Defaults to 0.

deephyper.contrib.callbacks.csv_extended_logger

deephyper.contrib.callbacks.learning_rate_warmup

Adapted from Horovod implementation: https://github.com/horovod/horovod/blob/master/horovod/keras/callbacks.py

deephyper.contrib.callbacks.stop_if_unfeasible

deephyper.contrib.callbacks.stop_on_timeout

deephyper.contrib.callbacks.time_stopping

Callback that stops training when a specified amount of time has passed.

deephyper.contrib.callbacks.utils