Can Keras with Tensorflow backend be forced to use CPU or GPU at will

Question

I have Keras installed with the Tensorflow backend and CUDA   I d like to sometimes on demand force Keras to use CPU   Can this be done without say installing a separate CPU-only Tensorflow in a virtual environment   If so how   If the backend were Theano  the flags could be set  but I have not heard of Tensorflow flags accessible via Keras

User · Accepted Answer

If you want to force Keras to use CPU

Way 1

import os
os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID"   # see issue #152
os.environ["CUDA_VISIBLE_DEVICES"] = ""

before Keras / Tensorflow is imported.

Way 2

Run your script as

$ CUDA_VISIBLE_DEVICES="" ./your_keras_code.py

See also

User · Answer

For people working on PyCharm  and for forcing CPU  you can add the following line in the Run Debug configuration  under Environment variables    lt OTHER ENVIRONMENT VARIABLES gt  CUDA VISIBLE DEVICES -1

User · Answer

A rather separable way of doing this is to use   import tensorflow as tf from keras import backend as K  num cores   4  if GPU      num GPU   1     num CPU   1 if CPU      num CPU   1     num GPU   0  config   tf ConfigProto intra op parallelism threads num cores                          inter op parallelism threads num cores                           allow soft placement True                          device count     CPU    num CPU                                           GPU    num GPU                            session   tf Session config config  K set session session    Here  with booleans GPU and CPU  we indicate whether we would like to run our code with the GPU or CPU by rigidly defining the number of GPUs and CPUs the Tensorflow session is allowed to access  The variables num GPU and num CPU define this value  num cores then sets the number of CPU cores available for usage via intra op parallelism threads and inter op parallelism threads   The intra op parallelism threads variable dictates the number of threads a parallel operation in a single node in the computation graph is allowed to use  intra   While the inter ops parallelism threads variable defines the number of threads accessible for parallel operations across the nodes of the computation graph  inter    allow soft placement allows for operations to be run on the CPU if any of the following criterion are met    there is no GPU implementation for the operation there are no GPU devices known or registered there is a need to co-locate with other inputs from the CPU   All of this is executed in the constructor of my class before any other operations  and is completely separable from any model or other code I use    Note  This requires tensorflow-gpu and cuda cudnn to be installed because the option is given to use a GPU   Refs    What do the options in ConfigProto like allow soft placement and log device placement mean  Meaning of inter op parallelism threads and intra op parallelism threads

User · Answer

As per keras tutorial  you can simply use the same tf device scope as in regular tensorflow   with tf device   gpu 0        x   tf placeholder tf float32  shape  None  20  64       y   LSTM 32  x     all ops in the LSTM layer will live on GPU 0  with tf device   cpu 0        x   tf placeholder tf float32  shape  None  20  64       y   LSTM 32  x     all ops in the LSTM layer will live on CPU 0

User · Answer

Just import tensortflow and use keras  it s that easy   import tensorflow as tf   your code here with tf device   gpu 0        model fit X  y  epochs 20  batch size 128  callbacks callbacks list

User · Answer

This worked for me  win10   place before you import keras   import os os environ  CUDA VISIBLE DEVICES      -1

User · Answer

I just spent some time figure it out  Thoma s answer is not complete  Say your program is test py  you want to use gpu0 to run this program  and keep other gpus free   You should write CUDA VISIBLE DEVICES 0 python test py  Notice it s DEVICES not DEVICE

[python] Can Keras with Tensorflow backend be forced to use CPU or GPU at will?

Way 1

Way 2

Examples related to python

Examples related to machine-learning

Examples related to tensorflow

Examples related to keras