我是否有可能只用cpu训练inception-v3模型？(Is it possible that I can train inception-v3 model only with cpu?)

 当我尝试使用我自己的基于Inception-v3的图像进行训练时如下： 
 
 https://github.com/tensorflow/models/blob/master/inception/inception/inception_train.py 
 
 我收到此错误： 
 
File "/tensorflow/tensorflow/models/inception/bazel-bin/inception/flowers_train.runfiles/inception/inception_train.py", line 207, in train
    assert FLAGS.batch_size % FLAGS.num_gpus == 0, (
ZeroDivisionError: division by zero
 
 我的问题是“我是否有可能只用cpu训练起始-v3模型？” 
 
 PS：我知道如果只使用cpu会比gpu花费更多时间，因此我减少了max_steps和图像数据。 
 
 非常感谢你 ！ 

When I try to train with my own image base on Inception-v3 as follow:
 
https://github.com/tensorflow/models/blob/master/inception/inception/inception_train.py
 
I get this error:
 
File "/tensorflow/tensorflow/models/inception/bazel-bin/inception/flowers_train.runfiles/inception/inception_train.py", line 207, in train
    assert FLAGS.batch_size % FLAGS.num_gpus == 0, (
ZeroDivisionError: division by zero
 
My question is "Is it possible that I can train inception-v3 model only with cpu ?"
 
PS: I know if only use cpu which will spend much time than gpu therefore, I have decrease the max_steps and image data.
 
Thank you very much !

原文：https://stackoverflow.com/questions/37230843