Login Register

tf.contrib.learn.monitors.NanLoss

class tf.contrib.learn.monitors.NanLoss

NaN Loss monitor.

Monitors loss and stops training if loss is NaN. Can either fail with exception or just stop training.

Links:

https://www.tensorflow.org/versions/master/api_docs/python/contrib.learn.monitors.html#NanLoss

doc_TensorFlow

2016-10-14 13:06:27

Comments

Leave a Comment

Please login to continue.

Popular Articles

tf.contrib.learn.monitors.NanLoss.begin() tf.contrib.learn.monitors.NanLoss.begin(max_steps=None) Called at the beginning of training. When called, the default graph is the one we are execut

tf.contrib.learn.monitors.NanLoss.__init__() tf.contrib.learn.monitors.NanLoss.__init__(loss_tensor, every_n_steps=100, fail_on_nan_loss=True) Initializes NanLoss monitor. Args: loss_tensor: T

tf.contrib.learn.monitors.NanLoss.end() tf.contrib.learn.monitors.NanLoss.end(session=None)

tf.contrib.learn.monitors.NanLoss.post_step() tf.contrib.learn.monitors.NanLoss.post_step(step, session)

tf.contrib.learn.monitors.NanLoss.epoch_begin() tf.contrib.learn.monitors.NanLoss.epoch_begin(epoch) Begin epoch. Args: epoch: int, the epoch number. Raises: ValueError: if we've already begun