这里写自定义目录标题
bert 连续输出loss
解决方法：
train_op = optimization.create_optimizer(
total_loss, learning_rate, num_train_steps, num_warmup_steps, use_tpu)
logging_hook = tf.train.LoggingTensorHook({“total_loss:”: total_loss}, every_n_iter=10)
output_spec = tf.contrib.tpu.TPUEstimatorSpec(
mode=mode,
loss=total_loss,
train_op=train_op,
training_hooks=[logging_hook],
scaffold_fn=scaffold_fn)
2762
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
