How To Use Your GPU for Machine Learning on Windows with Jupyter Notebook and Tensorflow

This video will show you how to use your GPU for machine learning. First, Install Anaconda Navigator. Next you will create a virtual environment for your python project. A virtual environment is like a self-contained custom directory for your python IDE. Name it what you want and select Python version 3.6 I’ve create a working version of the virtual environment with the necessary packages for using your GPU for machine learning. Here’s how to create one from scratch.

00:41 - The keras, tensorflow, and tensor-gpu libraries need to be installed within your python directories. Check what is not installed, select keras, tensorflow, and tensorflow-gpu to install the required packages. This will show you all packages and sub-packages that will be installed. Now install JupyterLab in this custom virtual environment. Both JupyterLab and Jupyter Notebook will be installed. Both are python integrated development environments for writing and testing your code. I personally like JupyterLab because of the quality of life features, such as using the mouse to rearrange functions and the displaying of all file system and directories for your project. It’s very well integrated for data visualization which helps make for a more dynamic programming experience. Launch JupyterLab. Your browser will open up a JupyterLab session. Run the test script to check if the GPU is being detected by python. The number of GPU reads as 1.

02:30 - You can now use your GPU to train your machine learning models. I hope this video was helpful. Like, Comment, and Subscribe. .