TensorFlow¶

Anaconda makes it easy to install TensorFlow, enabling your data science, machine learning, and artificial intelligence workflows.

This page shows how to install TensorFlow with the conda package manager included in Anaconda and Miniconda.

TensorFlow with conda is supported on 64-bit Windows 7 or later, 64-bit Ubuntu Linux 14.04 or later, 64-bit CentOS Linux 6 or later, and macOS 10.10 or later.

The instructions are the same for all operating systems. No apt install or yum install commands are required.

Install TensorFlow¶

Download and install Anaconda or the smaller Miniconda.
On Windows open the Start menu and open an Anaconda Command Prompt. On macOS or Linux open a terminal window. Use the default bash shell on macOS or Linux.
Choose a name for your TensorFlow environment, such as “tf”.
To install the current release of CPU-only TensorFlow, recommended for beginners:
```
conda create -n tf tensorflow
conda activate tf
```
Or, to install the current release of GPU TensorFlow on Linux or Windows:
```
conda create -n tf-gpu tensorflow-gpu
conda activate tf-gpu
```

TensorFlow is now installed and ready to use.

For using TensorFlow with a GPU, refer to the TensorFlow documentation on the topic, specifically the section on device placement.

CUDA versions¶

GPU TensorFlow uses CUDA. On Windows and Linux only CUDA 10.0 is supported for the TensorFlow 2.0 release. Previous versions of TensorFlow support other version of CUDA.

To install GPU TensorFlow with a non-default CUDA version such as 9.0 run:

conda create -n tf-gpu-cuda8 tensorflow-gpu cudatoolkit=9.0
conda activate tf-gpu-cuda8

Python 2¶

Anaconda recommends Python 3, but it is possible to use TensorFlow with Python 2 on Linux and macOS.

CPU-only TensorFlow on Python 2 on Linux or macOS:

conda create -n tf-2 tensorflow python=2
conda activate tf-2

GPU TensorFlow on Python 2 on Linux:

conda create -n tf-2-gpu tensorflow python=2
conda activate tf-2-gpu

Nightly builds¶

Advanced users may wish to install the latest nightly build of TensorFlow. These nightly builds are unstable and are only available as pip packages on PyPI.

To install the nightly build of CPU-only TensorFlow:

conda create -n tf-n python
conda activate tf-n
pip install tf-nightly

Or, to install the nightly build of GPU TensorFlow on Linux or Windows:

conda create -n tf-n-gpu python
conda activate tf-n-gpu
pip install tf-nightly-gpu

Nightly build of CPU-only TensorFlow on Python 2 on Linux or macOS:

conda create -n tf-2-n python=2
conda activate tf-2-n
pip install tf-nightly

Nightly build of GPU TensorFlow on Python 2 on Linux:

conda create -n tf-2-n-gpu python=2
conda activate tf-2-n-gpu
pip install tf-nightly-gpu