Installing and Running Tensorflow-GPU using Anaconda on Windows
How to install Tensorflow-GPU on Windows 10
Requirements; Anaconda, A PC with Windows 10 OS, Internet connection and at least 1GB of data.
1. Run Anaconda
If you do not have Anaconda installed, you can download it from here and install then search for Anaconda in the Windows search bar. Run the Navigator and Anaconda prompt.
2. Create an Environment
You can create a new environment by typing the following command on Anaconda Prompt.
conda create -n gputensorflow python=3.7
For the purpose of this article, the environment name is gputensorflow and we’re installing python version 3.7, you can choose any environment name.
3. Activate the Environment
conda activate gputensorflow
4. Install the ipykernel
To install the kernel, enter the following command;
pip install ipykernel
python –m ipykernel install --user --name gputensorflow --display-name “gputensorflow”
5. Installing Tensorflow-GPU
conda install tensorflow-gpu
Anaconda installs Tensorflow-GPU and a number of libraries. This may take a longer time than other installations.
6. Restart your PC (optional)
7. Run Anaconda and the TensorFlow environment
When you open the Anaconda Navigator, click on the arrow beside the “Applications on” and click on your environment.
8. Launch Jupyter Notebook
You should have something like this 👇
There you go, you can now create your Neural Networks and other stuff with TensorFlow. You can also install other libraries.
I hope the article was helpful, If you encountered any challenge while following the steps, you can send a Direct Message to me on Twitter or drop a comment in the Response section.