TensorFlow is an open-source software library developed by Google for machine learning and deep learning tasks. It provides a flexible and efficient way to build and train machine learning models, from simple linear regression to complex neural networks. TensorFlow supports a wide range of platforms, including Ubuntu, one of the most popular Linux distributions.

If you’re new to TensorFlow and want to get started with Ubuntu, this step-by-step tutorial will guide you through the process of installing TensorFlow on your Ubuntu system. By the end of this tutorial, you’ll have a working installation of TensorFlow and be ready to start building and training your own machine-learning models.

Hardware Requirements

The hardware requirements for installing TensorFlow on Ubuntu depend on the type of installation you are doing, specifically, whether you are installing the CPU version or the GPU version of TensorFlow. Here are the basic hardware requirements for each version:

  • CPU version
    • Processor: x86-64 CPU with AVX2 support
    • RAM: 2GB or more
    • Disk space: 500MB or more
  • GPU version
    • Processor: NVIDIA GPU with compute capability of 3.5 or higher (see the list of supported GPUs)
    • NVIDIA CUDA Toolkit: version 11.0 or higher (see the CUDA Toolkit requirements)
    • cuDNN library: version 8.0 or higher (see the cuDNN requirements)
    • RAM: 4GB or more
    • Disk space: 500MB or more

It’s also worth noting that using a GPU can significantly speed up the training of deep learning models, so if you have a compatible NVIDIA GPU, it’s highly recommended to install the GPU version of TensorFlow. However, the CPU version can still be used for many machine-learning tasks and is a good option if you don’t have a compatible GPU.

Step 1: Update your system

Before we get started, it’s important to make sure your Ubuntu system is up to date. Open a terminal window and run the following command to update your system’s package list and install any available updates:

sudo apt update && sudo apt upgrade 

Step 2: Install Python 3 and PIP

TensorFlow requires Python 3.7-3.10, so if you don’t have Python installed, you can install it by running:

sudo apt install python3-dev python3-pip 

This will also install pip, the package installer for Python.

Step 3: Create a virtual environment (optional)

It’s recommended to use a virtual environment to isolate TensorFlow and its dependencies from other packages on your system. First install the Python package for creating virtual environments:

sudo apt install python3-venv 

Next, create a new directory for your Tensorflow application and switch to them:

mkdir tensorflow-app && cd tensorflow-app 

Then, you can create a virtual environment by running:

python3 -m venv venv

This will create a virtual environment named “venv” in the current directory.

To activate the virtual environment, run:

source venv/bin/activate 

You should now see the virtual environment name in your terminal prompt.

Step 4: Install TensorFlow

With Python and pip installed, you can now install TensorFlow. To install the CPU version of TensorFlow, run:

pip install --upgrade tensorflow 

If you have a GPU and want to install the GPU version of TensorFlow, you’ll need to install the NVIDIA CUDA Toolkit and cuDNN library first. You can find instructions for installing these libraries in the TensorFlow documentation.

Step 5: Install TensorFlow Addons (optional)

TensorFlow Addons is a repository of community-developed extensions and plugins for TensorFlow. To install TensorFlow Addons, run:

pip install --upgrade tensorflow-addons 

This will install the latest version of TensorFlow Addons.

Step 6: Test the installation

Once TensorFlow is installed, you can test the installation by running a simple script that creates and runs a TensorFlow session:

python -c "import tensorflow as tf; print(tf.version.VERSION)" 

If TensorFlow is installed correctly, you should see the version of the TensorFlow printed to the terminal.

Step 7: Install Jupyter Notebook (optional)

Jupyter Notebook is a web-based interactive development environment for Python. It’s a great tool for exploring and experimenting with TensorFlow code. To install Jupyter Notebook, run:

pip install jupyter 

To launch Jupyter Notebook, run:

jupyter notebook 

This will start the Jupyter Notebook server and open a new browser window with the Jupyter Notebook interface.

Step 8: Create a new notebook (optional)

In the Jupyter Notebook interface, click “New” and select “Python 3” to create a new notebook. In the first cell, type the following code to import TensorFlow:

You can now start experimenting with TensorFlow in your Jupyter Notebook!

Step 9: Deactivate the virtual environment (optional)

If you created a virtual environment in step 3, you can deactivate it by running:

deactivate 

This will return you to your system’s default Python environment.

And that’s it! You should now have TensorFlow installed and ready to use on your Ubuntu system. Whether you’re new to machine learning or an experienced developer, TensorFlow on Ubuntu is a powerful tool for building and training machine learning models.

Conclusion

Installing TensorFlow on Ubuntu can be a bit daunting, especially if you’re new to machine learning and Linux. However, with this step-by-step tutorial, you should be able to install TensorFlow with ease and get started with machine learning on Ubuntu.

Remember to keep your system up to date, create a virtual environment for TensorFlow, and test the installation to make sure everything is working correctly. Once you’ve installed TensorFlow, the possibilities are endless, and you’ll have a powerful tool for building and training your own machine-learning models.

Whether you’re a student, researcher, or developer, TensorFlow on Ubuntu is a great way to explore the world of machine learning and unlock the potential of this exciting field. So, what are you waiting for? Start your journey with TensorFlow on Ubuntu today!