Python is increasingly becoming a popular programming language for machine learning and deep learning. If you want to use python for training a deep neural network, GPU is preferable rather CPU. This article explains the setting up python environment for deep learning (GPU), especially using
Anaconda.
Anaconda is a python distribution for scientific computing and machine learning. Follow the below steps to set up the environment.
1. Install Anaconda
Anaconda distribution is available for Windows, Linux, and Mac operating systems. Download the suitable package using the following link :
Download Anaconda. After downloading the anaconda distribution, you just execute the setup file and follow the instructions in the wizard.
2. Update Anaconda
After installing Anaconda, open anaconda prompt. (go to start -> search for anaconda prompt as shown below.)
Anaconda prompt looks like command prompt. there execute following two commands. (see the below figure)
conda update conda
conda update --all
|
3. CUDA Tool kit and CuDNN installation
After successful updation of Anaconda distribution, you have to install two softwares : (i) CUDA Tool Kit (2) CuDNN.
It is note that you have to install compatible softwares based on your OS and GPU. Please follow the links for more information.
CUDA toolkit
Download CuDNN
4. CuDNN path setting using environment variable
