Implementation-of-TensorFlow-GPU-CUDA-in-Windows

Documentation: Implementation of TensorFlow GPU (CUDA) in Windows

Environment Setup

1. Download Anaconda from https://www.anaconda.com/products/individual and Python from https://www.python.org/downloads/. Once Anaconda has been downloaded, run the command conda install -y jupyter
   #Suggestion: Remember to TICK set-up the environment path during the installation process of Anaconda

2. To set-up the TensorFlow GPU (CUDA) and its compatible environment, please refer to this YouTube link https://www.youtube.com/watch?v=qrkEYf-YDyI&list=PLjy4p-07OYzulelvJ5KVaT2pDlxivl_BN.

3. Copy the script from https://github.com/JJLim99/Implementation-of-TensorFlow-GPU-CUDA-in-Windows/blob/master/gpu.yml and save it as gpu.yml. (Remember where you save this file)

4. Open cmd, go to the file location where you save gpu.yml. Then, run the following command:
conda env create -v -f gpu.yml
conda activate gpu
python -m ipykernel install --user --name gpu --display-name "Python (GPU)"

# To test your environment, copy the script from https://github.com/JJLim99/Implementation-of-TensorFlow-GPU-CUDA-in-Windows/blob/master/Version.ipynb and paste it to your new python script in jupyter notebook. (Remember that you must in the tensorflow environment)