tensorflow中tf.contrib.layers.optimize_loss的参数optimizer使用

深度学习优化器的多种使用方式

最新推荐文章于 2021-12-14 19:59:37 发布

原创

最新推荐文章于 2021-12-14 19:59:37 发布 · 1k 阅读

0 ·

CC 4.0 BY-SA版权

文章标签：

#tensorflow

def optimize_loss(loss,
                  global_step,
                  learning_rate,
                  optimizer,
                  gradient_noise_scale=None,
                  gradient_multipliers=None,
                  clip_gradients=None,
                  learning_rate_decay_fn=None,
                  update_ops=None,
                  variables=None,
                  name=None,
                  summaries=None,
                  colocate_gradients_with_ops=False,
                  increment_global_step=True):

optimizer的多种传参方式：

- by string specifying the name of the optimizer. See OPTIMIZER_CLS_NAMES
    for full list. E.g. `optimize_loss(..., optimizer='Adam')`.
- by function taking learning rate `Tensor` as argument and returning an
    `Optimizer` instance. E.g. `optimize_loss(...,
    optimizer=lambda lr: tf.train.MomentumOptimizer(lr, momentum=0.5))`.
  Alternatively, if `learning_rate` is `None`, the