How to Check If Keras Is Using Gpu Version of Tensorflow

How do I check if keras is using gpu version of tensorflow?

You are using the GPU version. You can list the available tensorflow devices with (also check this question):

from tensorflow.python.client import device_lib
print(device_lib.list_local_devices()) # list of DeviceAttributes

EDIT:

With tensorflow >= 1.4 you can run the following function:

import tensorflow as tf
tf.test.is_gpu_available() # True/False

# Or only check for gpu's with cuda support
tf.test.is_gpu_available(cuda_only=True) 

EDIT 2:

The above function is deprecated in tensorflow > 2.1. Instead you should use the following function:

import tensorflow as tf
tf.config.list_physical_devices('GPU')

---

NOTE:

In your case both the cpu and gpu are available, if you use the cpu version of tensorflow the gpu will not be listed. In your case, without setting your tensorflow device (with tf.device("..")), tensorflow will automatically pick your gpu!

In addition, your sudo pip3 list clearly shows you are using tensorflow-gpu. If you would have the tensoflow cpu version the name would be something like tensorflow(1.1.0).

Check this issue for information about the warnings.

How to Know if Keras is using GPU or CPU

First lets make sure tensorflow is detecting your GPU. Run the code below. If number of GPUs=0 it is not detecting your GPU. For tensorflow to use the GPU you need to have the Cuda toolkit and Cudnn installed. If no GPU is detected and you are using Anaconda reinstall tensorflow with Conda. It automatically installs the toolkit and Cudnn. Pip does not install these when you use it to install tensorflow.

import tensorflow as tf
from tensorflow.python.client import device_lib
print(device_lib.list_local_devices())
print(tf.__version__)
print("Num GPUs Available: ", len(tf.config.experimental.list_physical_devices('GPU')))
tf.test.is_gpu_available()
!python --version

Check whether Tensorflow is running on GPU

Roberto, try this

import tensorflow as tf
if tf.test.gpu_device_name():
    print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))
else:
    print("Please install GPU version of TF")

the output should be

Default GPU Device: /device:GPU:0

GPU 0 is your GTX 860m

How ensure that Keras is using GPU with tensorflow backend?

You need to run your network with log_device_placement = True set in the TensorFlow session (the line before the last in the sample code below.) Interestingly enough, if you set that in a session, it will still apply when Keras does the fitting. So this code below (tested) does output the placement for each tensor. Please note, I've short-circuited the data reading because your data wan't available, so I'm just running the network with random data. The code this way is self-contained and runnable by anyone. Another note: if you run this from Jupyter Notebook, the output of the log_device_placement will go to the terminal where Jupyter Notebook was started, not the notebook cell's output.

from tensorflow.contrib.keras.api.keras.models import Sequential
from tensorflow.contrib.keras.api.keras.layers import Dense
from tensorflow.contrib.keras.api.keras.layers import Dropout
from tensorflow.contrib.keras.api.keras import utils as np_utils
import numpy as np
import pandas as pd
import tensorflow as tf

# Read data
#pddata=pd.read_csv('data/data.csv', delimiter=';')
pddata = "foobar"

# Helper function (prepare & test data)
def split_to_train_test (data):

    return (
        np.random.uniform( size = ( 100, 101 ) ),
        np.random.uniform( size = ( 100, 101 ) ),
        np.random.randint( 0, size = ( 100 ), high = 3 ),
        np.random.randint( 0, size = ( 100 ), high = 3 )
    )

    trainLenght = len(data) - len(data)//10

    trainData = data.loc[:trainLenght].sample(frac=1).reset_index(drop=True)
    testData = data.loc[trainLenght+1:].sample(frac=1).reset_index(drop=True)

    trainLabels = trainData.loc[:,"Label"].as_matrix()
    testLabels = testData.loc[:,"Label"].as_matrix()

    trainData = trainData.loc[:,"Feature 0":].as_matrix()
    testData  = testData.loc[:,"Feature 0":].as_matrix()

    return (trainData, testData, trainLabels, testLabels)

# prepare train & test data
(X_train, X_test, y_train, y_test) = split_to_train_test (pddata)

# Convert labels to one-hot notation
Y_train = np_utils.to_categorical(y_train, 3)
Y_test  = np_utils.to_categorical(y_test, 3)

# Define model in Keras
def create_model(init):
    model = Sequential()
    model.add(Dense(101, input_shape=(101,), kernel_initializer=init, activation='tanh'))
    model.add(Dense(101, kernel_initializer=init, activation='tanh'))
    model.add(Dense(101, kernel_initializer=init, activation='tanh'))
    model.add(Dense(101, kernel_initializer=init, activation='tanh'))
    model.add(Dense(3, kernel_initializer=init, activation='softmax'))
    return model

# Train the model
uniform_model = create_model("glorot_normal")
uniform_model.compile(loss='categorical_crossentropy', optimizer='sgd', metrics=['accuracy'])
with tf.Session( config = tf.ConfigProto( log_device_placement = True ) ):
    uniform_model.fit(X_train, Y_train, batch_size=1, epochs=300, verbose=1, validation_data=(X_test, Y_test)) 

Terminal output (partial, it was way too long):

...

VarIsInitializedOp_13: (VarIsInitializedOp): /job:localhost/replica:0/task:0/device:GPU:0

2018-04-21 21:54:33.485870: I tensorflow/core/common_runtime/placer.cc:884]

VarIsInitializedOp_13: (VarIsInitializedOp)/job:localhost/replica:0/task:0/device:GPU:0

training/SGD/mul_18/ReadVariableOp: (ReadVariableOp): /job:localhost/replica:0/task:0/device:GPU:0

2018-04-21 21:54:33.485895: I tensorflow/core/common_runtime/placer.cc:884]

training/SGD/mul_18/ReadVariableOp: (ReadVariableOp)/job:localhost/replica:0/task:0/device:GPU:0

training/SGD/Variable_9/Read/ReadVariableOp: (ReadVariableOp): /job:localhost/replica:0/task:0/device:GPU:0

2018-04-21 21:54:33.485903: I tensorflow/core/common_runtime/placer.cc:884]

training/SGD/Variable_9/Read/ReadVariableOp: (ReadVariableOp)/job:localhost/replica:0/task:0/device:GPU:0

...

Note the GPU:0 at the end of many lines.

Tensorflow manual's relevant page: Using GPU: Logging Device Placement.

---

Related Topics