为什么`tf.nn.nce_损失`在tensorflow不能在GPU上运行？

Question

代码tensorflow/examples/tutorials/word2vec/word2vec_basic.py有一个注释# Ops and variables pinned to the CPU because of missing GPU implementation。我还发现操作tf.nn.nce_loss不能由GPU实现。那为什么呢tf.nn.nce_损失无法在GPU上运行？在

我使用log_device_placement来查看MUST BE CPU操作。结果如下：

nce_loss/LogUniformCandidateSampler: (LogUniformCandidateSampler): /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/LogUniformCandidateSampler: (LogUniformCandidateSampler)/job:localhost/replica:0/task:0/cpu:0
nce_loss/ComputeAccidentalHits: (ComputeAccidentalHits): /job:localhost/replica:0/task:0cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/ComputeAccidentalHits: (ComputeAccidentalHits)/job:localhost/replica:0/task:0/cpu:0
nce_loss/SparseToDense: (SparseToDense): /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/SparseToDense: (SparseToDense)/job:localhost/replica:0/task:0/cpu:0
nce_loss/concat: (ConcatV2): /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/concat: (ConcatV2)/job:localhost/replica:0/task:0/cpu:0
nce_loss/LogUniformCandidateSampler: (LogUniformCandidateSampler): /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/LogUniformCandidateSampler: (LogUniformCandidateSampler)/job:localhost/replica:0/task:0/cpu:0
nce_loss/ComputeAccidentalHits: (ComputeAccidentalHits): /job:localhost/replica:0/task:0cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/ComputeAccidentalHits: (ComputeAccidentalHits)/job:localhost/replica:0/task:0/cpu:0
nce_loss/SparseToDense: (SparseToDense): /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/SparseToDense: (SparseToDense)/job:localhost/replica:0/task:0/cpu:0
nce_loss/concat: (ConcatV2): /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/concat: (ConcatV2)/job:localhost/replica:0/task:0/cpu:0
nce_loss/LogUniformCandidateSampler: (LogUniformCandidateSampler): /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/LogUniformCandidateSampler: (LogUniformCandidateSampler)/job:localhost/replica:0/task:0/cpu:0
nce_loss/ComputeAccidentalHits: (ComputeAccidentalHits): /job:localhost/replica:0/task:0cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/ComputeAccidentalHits: (ComputeAccidentalHits)/job:localhost/replica:0/task:0/cpu:0
nce_loss/SparseToDense: (SparseToDense): /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/SparseToDense: (SparseToDense)/job:localhost/replica:0/task:0/cpu:0
nce_loss/concat: (ConcatV2): /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/concat: (ConcatV2)/job:localhost/replica:0/task:0/cpu:0
nce_loss/LogUniformCandidateSampler: (LogUniformCandidateSampler): /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/LogUniformCandidateSampler: (LogUniformCandidateSampler)/job:localhost/replica:0/task:0/cpu:0
nce_loss/ComputeAccidentalHits: (ComputeAccidentalHits): /job:localhost/replica:0/task:0cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/ComputeAccidentalHits: (ComputeAccidentalHits)/job:localhost/replica:0/task:0/cpu:0
nce_loss/SparseToDense: (SparseToDense): /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/SparseToDense: (SparseToDense)/job:localhost/replica:0/task:0/cpu:0
nce_loss/concat: (ConcatV2): /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/concat: (ConcatV2)/job:localhost/replica:0/task:0/cpu:0
nce_loss/LogUniformCandidateSampler: (LogUniformCandidateSampler): /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/LogUniformCandidateSampler: (LogUniformCandidateSampler)/job:localhost/replica:0/task:0/cpu:0
nce_loss/ComputeAccidentalHits: (ComputeAccidentalHits): /job:localhost/replica:0/task:0cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/ComputeAccidentalHits: (ComputeAccidentalHits)/job:localhost/replica:0/task:0/cpu:0
nce_loss/SparseToDense: (SparseToDense): /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/SparseToDense: (SparseToDense)/job:localhost/replica:0/task:0/cpu:0
nce_loss/concat: (ConcatV2): /job:localhost/replica:0/task:0/cpu:0
I tensorflow/core/common_runtime/simple_placer.cc:841] nce_loss/concat: (ConcatV2)/job:localhost/replica:0/task:0/cpu:0

但是，我还是不知道是哪个特种部队造成了这个问题。是否有GPU无法执行的特殊操作？或者这是由于tensorflow本身缺乏实现？在

Answer 1

您也可以使用gpu accelerator执行此示例。
请参见tensorflow支持的设备https://www.tensorflow.org/guide/using_gpu#supported_devices
您必须添加以下代码行

with tf.device('/gpu:0'):

并删除#with tf.device('/cpu:0'):

基本上在这条线上用gpu替换cpu： https://github.com/tensorflow/tensorflow/blob/bd965956535b1026a0769d04c3f6419d020f5f9c/tensorflow/examples/tutorials/word2vec/word2vec_basic.py#L189

希望这有帮助。在