tensorflow (version 2.4.0)

install_tensorflow: Install TensorFlow and its dependencies


Install TensorFlow and its dependencies


  method = c("auto", "virtualenv", "conda"),
  conda = "auto",
  version = "default",
  envname = NULL,
  extra_packages = NULL,
  restart_session = TRUE,
  conda_python_version = "3.7",



Installation method. By default, "auto" automatically finds a method that will work in the local environment. Change the default to force a specific installation method. Note that the "virtualenv" method is not available on Windows (as this isn't supported by TensorFlow). Note also that since this command runs without privilege the "system" method is available only on Windows.


The path to a conda executable. Use "auto" to allow reticulate to automatically find an appropriate conda binary. See Finding Conda for more details.


TensorFlow version to install. Up to and including TensorFlow 2.0, specify "default" to install the CPU version of the latest release; specify "gpu" to install the GPU version of the latest release. Starting from TensorFlow 2.1, by default a version is installed that works on both GPU- and CPU-only systems. Specify "cpu" to install a CPU-only version.

You can also provide a full major.minor.patch specification (e.g. "1.1.0"), appending "-gpu" if you want the GPU version (e.g. "1.1.0-gpu").

Alternatively, you can provide the full URL to an installer binary (e.g. for a nightly binary).


Name of Python environment to install within


Additional Python packages to install along with TensorFlow.


Restart R session after installing (note this will only occur within RStudio).


the python version installed in the created conda environment. Python 3.6 is installed by default.