Keras Implement of Lazy Optimizer

Inheriting Optimizer class, wrapping the original optimizer to achieve a new corresponding lazy optimizer.

Here we use gradients are equal to zeros or not to distinguish whether the words are sampled or not.

Usage

just replace your original with-momentum optimizer, like Adam(1e-3), with LazyOptimizer(Adam(1e-3), embedding_layers).

see imdb_lstm_test.py.

Lazy类优化器的Keras实现

继承Optimizer类，包装原有优化器，实现Lazy版优化器。

这里判断一个词是否被采样的方法是检查该词的梯度是否全为0。

用法

直接将原来用的带动量的优化器, 如 Adam(1e-3), 替换为 LazyOptimizer(Adam(1e-3), embedding_layers) 就行了.

参考 imdb_lstm_test.py.

交流

QQ交流群：67729435，微信群请加机器人微信号spaces_ac_cn