Tensorflow 2.0中是否有cudnnLSTM或cudNNGRU替代品 [英] Is there cudnnLSTM or cudNNGRU alternative in tensorflow 2.0

问题描述

TensorFlow 1.0中的CuDNNGRU确实非常快.但是当我转到TensorFlow 2.0时，我找不到CuDNNGRU.简单的GRU在TensorFlow 2.0中真的很慢.

The CuDNNGRU in TensorFlow 1.0 is really fast. But when I shifted to TensorFlow 2.0 i am unable to find CuDNNGRU. Simple GRU is really slow in TensorFlow 2.0.

有什么方法可以在TensorFlow 2.0中使用CuDNNGRU?

Is there any way to use CuDNNGRU in TensorFlow 2.0?

推荐答案

可导入的实现已被弃用-相反，如果所有

The importable implementations have been deprecated - instead, LSTM and GRU will default to CuDNNLSTM and CuDNNGRU if all conditions are met:

1. activation = 'tanh'
2. recurrent_activation = 'sigmoid'
3. recurrent_dropout = 0
4. unroll = False
5. use_bias = True
6. 输入如果被屏蔽，请严格右击
7. reset_after = True(仅限GRU)

1. activation = 'tanh'
2. recurrent_activation = 'sigmoid'
3. recurrent_dropout = 0
4. unroll = False
5. use_bias = True
6. Inputs, if masked, are strictly right-padded
7. reset_after = True (GRU only)

---

还要确保TensorFlow使用GPU:

---

Also ensure TensorFlow uses GPU:

import tensorflow as tf
from tensorflow.python.client import device_lib
print(device_lib.list_local_devices())
print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))

---

更新:在Colab上运行时，似乎在TF 2.0.0上存在问题，以使CuDNN正常工作；尝试使用!pip install tensorflow==2.1.0.

---

Update: there appears to be a problem w/ TF 2.0.0 when running on Colab in getting CuDNN to work; try !pip install tensorflow==2.1.0 instead.

这篇关于Tensorflow 2.0中是否有cudnnLSTM或cudNNGRU替代品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！