In this tutorial, you will learn to install TensorFlow 2.0 on your macOS system running either Catalina or Mojave<\/p>\n There are a number of important updates in TensorFlow 2.0, including eager execution, automatic differentiation, and better multi-GPU\/distributed training support,\u00a0but the\u00a0most important update<\/em> is that Keras is now the\u00a0official<\/em> high-level deep learning API for TensorFlow.<\/strong><\/p>\n Furthermore, if you own a copy of my book,\u00a0Deep Learning for Computer Vision with Python<\/em><\/a>, you should use this guide to properly install TensorFlow 2.0 on your macOS system.<\/p>\n Inside this tutorial, you\u2019ll learn how to install TensorFlow 2.0 on macOS (using either Catalina or Mojave).<\/p>\n Alternatively, click here for my Ubuntu + TensorFlow 2.0<\/strong> installation instructions<\/a>.<\/p>\n To learn how to install TensorFlow 2.0 on macOS, just keep reading.<\/em><\/strong><\/p>\n In the first part of this tutorial, we\u2019ll briefly discuss the pre-configured deep learning development environments that are a part of my book, Deep Learning for Computer Vision with Python<\/a>.<\/em><\/p>\n We\u2019ll then configure and install TensorFlow 2.0 on our macOS system.<\/p>\n Let\u2019s begin.<\/p>\n Figure 1:<\/strong> My deep learning Virtual Machine with TensorFlow, Keras, OpenCV, and all other Deep Learning and Computer Vision libraries you need, pre-configured and pre-installed.<\/p>\n<\/div>\n When it comes to working with deep learning and Python I highly recommend<\/em> that you use a Unix-based environment.<\/p>\n Deep learning tools can be more easily configured and installed on Unix systems, allowing you to develop and run neural networks quickly.<\/p>\n Of course, configuring your own deep learning + Python + Linux development environment can be quite the tedious task, especially<\/em> if you are new to Unix, a beginner at working the command line\/terminal, or a novice when compiling and installing packages by hand.<\/p>\n In order to help you jump start your deep learning + Python education, I have created two pre-configured environments:<\/p>\n I strongly urge<\/em><\/strong> you to consider using my pre-configured environments if you are working through my books. Using a pre-configured environment is not cheating \u2014\u00a0 they simply allow you to focus on learning rather than the job of a system administrator.<\/strong><\/p>\n If you are more familiar with Microsoft Azure\u2019s infrastructure, be sure to check out their Ubuntu Data Science Virtual Machine (DSVM), including my review of the environment<\/a>. The Azure team maintains a great environment for you and I cannot speak highly enough about the support they provided while I ensured that all of my deep learning chapters ran successfully on their system.<\/p>\n That said, pre-configured environments are not for everyone.<\/p>\n In the remainder of this tutorial, we will serve as the \u201cdeep learning systems administrators\u201d installing TensorFlow 2.0 on our bare metal macOS machine.<\/p>\n The following instructions for installing TensorFlow 2.0 on your machine assume either:<\/p>\n Let\u2019s get started!<\/p>\n Figure 3:<\/strong> This tutorial supports installing TensorFlow 2.0 on macOS Mojave or macOS Catalina.<\/p>\n<\/div>\n These TensorFlow 2.0 install instructions in this guide are compatible with the following operating systems:<\/p>\n Alternatively, you may follow my Ubuntu + TensorFlow 2.0 installation instructions<\/a>.<\/p>\n Note:<\/strong> You may be wondering \u201cWhat about Windows?\u201d Keep in mind I do not support Windows on the PyImageSearch blog. You can read more about my \u201cno Windows policy\u201d in my FAQ<\/a>.<\/em><\/p>\n This step is for macOS Catalina only<\/em><\/strong>. If you use Mojave, please ignore this step and skip to\u00a0Step #3<\/strong>.<\/p>\n On macOS Catalina, you have the option of using a Bash shell or ZSH shell.<\/p>\n Bash is probably what you\u2019re used to.<\/strong> Previous versions of macOS have used Bash and by default, Ubuntu uses Bash as well. Apple has now made a decision that going forward, their operating systems will use ZSH.<\/strong><\/p>\n So what\u2019s the big deal and should I switch?<\/strong><\/em><\/p>\n Describing the differences between Bash and ZSH is outside the scope of this post \u2014 it\u2019s entirely up to you to decide. I would recommend reading this ZSH vs. Bash tutorial<\/a> to get you started, but again, the choice is up to you.<\/p>\n On one hand, you may discover that the changes do not impact you much and that using ZSH will be very comparable to Bash. Power users, on the other hand, may notice additional features.<\/p>\n If you\u00a0upgraded<\/em> from\u00a0High Sierra or Mojave to Catalina, by default your system will likely use Bash unless you explicitly change it by going into your terminal profile settings.<\/p>\n If your computer came with Catalina installed or you installed Catalina from scratch<\/em>, then your system will likely use ZSH by default.<\/p>\n Either way, if you decide to need make the switch on Catalina, you can follow these instructions to set your profile to use ZSH<\/a>.<\/p>\n Changing your shell is relatively simple via your \u201cTerminal Preferences\u201d > \u201cProfiles\u201d > \u201cShell\u201d<\/em> menu as shown in Figure 4<\/strong>:<\/p>\n![\"\"](\"https:\/\/www.pyimagesearch.com\/wp-content\/uploads\/2019\/12\/tensorflow2_install_macos_header.png\")
<\/a><\/p>\nHow to install TensorFlow 2.0 on macOS<\/h2>\n
Pre-configured deep learning environments<\/h3>\n
<\/p>\n\n
Configuring your macOS TensorFlow 2.0 deep learning system<\/h3>\n
\n
Step #1: Choose your macOS deep learning platform \u2014 either Catalina or Mojave<\/h4>\n
<\/a><\/p>\n\n
Step #2 (Catalina only): Choose Bash or ZSH as your shell<\/h4>\n