TensorFlow

Activating TensorFlow

This tutorial shows how to activate TensorFlow on an instance running the Deep Learning AMI with Conda (DLAMI on Conda) and run a TensorFlow program.

When a stable Conda package of a framework is released, it's tested and pre-installed on the DLAMI. If you want to run the latest, untested nightly build, you can Install TensorFlow's Nightly Build (experimental) manually.

To run TensorFlow on the DLAMI with Conda

1. To activate TensorFlow, open an Amazon Elastic Compute Cloud (Amazon EC2) instance of the DLAMI with Conda.

   • For TensorFlow and Keras 2 on Python 3 with CUDA 9.0 and MKL-DNN, run this command:

     $ source activate tensorflow_p36

   • For TensorFlow and Keras 2 on Python 2 with CUDA 9.0 and MKL-DNN, run this command:

     $ source activate tensorflow_p27

2. Start the iPython terminal:

   (tensorflow_p36)$ ipython

3. Run a TensorFlow program to verify that it is working properly:

   
   
   import tensorflow as tf
   hello = tf.constant('Hello, TensorFlow!')
   sess = tf.Session()
   print(sess.run(hello))
   

   Hello, TensorFlow! should appear on your screen.

Install TensorFlow's Nightly Build (experimental)

You can install the latest TensorFlow build into either or both of the TensorFlow Conda environments on your Deep Learning AMI with Conda.

To install TensorFlow from a nightly build

1. • For the Python 3 TensorFlow environment, run the following command:

     $ source activate tensorflow_p36

   • For the Python 2 TensorFlow environment, run the following command:

     $ source activate tensorflow_p27

2. Remove the currently installed TensorFlow.

   Note

   The remaining steps assume you are using the tensorflow_p36 environment.

   (tensorflow_p36)$ pip uninstall tensorflow

3. Install the latest nightly build of TensorFlow.

   (tensorflow_p36)$ pip install tf-nightly

4. To verify you have successfully installed latest nightly build, start the IPython terminal and check the version of TensorFlow.

   (tensorflow_p36)$ ipython

   import tensorflow
   print (tensorflow.__version__)

   The output should print something similar to 1.12.0-dev20181012

More Tutorials

TensorFlow with Horovod

TensorBoard

TensorFlow Serving

For tutorials, see the folder called Deep Learning AMI with Conda tutorials in the home directory of the DLAMI.

For more tutorials and examples, see the TensorFlow documentation for the TensorFlow Python API or see the TensorFlow website.