Keras:如何将学习率输出到张量板上 [英] Keras: how to output learning rate onto tensorboard

问题描述

我添加了一个回调来降低学习速度:

I add a callback to decay learning rate:

 keras.callbacks.ReduceLROnPlateau(monitor='val_loss', factor=0.5, patience=100, 
                                   verbose=0, mode='auto',epsilon=0.00002, cooldown=20, min_lr=0)

这是我的张量板回调:

keras.callbacks.TensorBoard(log_dir='./graph/rank{}'.format(hvd.rank()), histogram_freq=10, batch_size=FLAGS.batch_size,
                            write_graph=True, write_grads=True, write_images=False)

我想确保它在我的训练中开始生效，所以我想将学习率输出到tensorbaord.但是我找不到可以设置它的地方.

I want to make sure it have kicked in during my training, So I want to output learning rate onto tensorbaord.But I can not find where I can set it.

我还检查了优化程序的api，但没有运气.

I also checked optimizer api, but no luck.

keras.optimizers.Adam(lr=0.001, beta_1=0.9, beta_2=0.999, epsilon=None, decay=0.0, amsgrad=False)

那么我如何向Tensorboad输出学习率?

So How can I output learning rate to tensorboad?

推荐答案

根据Keras的作者，正确的方法是将TensorBoard回调子类化:

According to the author of Keras, the proper way is to subclass the TensorBoard callback:

from keras import backend as K
from keras.callbacks import TensorBoard

class LRTensorBoard(TensorBoard):
    def __init__(self, log_dir, **kwargs):  # add other arguments to __init__ if you need
        super().__init__(log_dir=log_dir, **kwargs)

    def on_epoch_end(self, epoch, logs=None):
        logs = logs or {}
        logs.update({'lr': K.eval(self.model.optimizer.lr)})
        super().on_epoch_end(epoch, logs)

然后将其作为callbacks参数的一部分传递给model.fit(贷方

Then pass it as part of the callbacks argument to model.fit (credit Finncent Price):

model.fit(x=..., y=..., callbacks=[LRTensorBoard(log_dir="/tmp/tb_log")])

这篇关于Keras:如何将学习率输出到张量板上的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！