Setting Up Your Deep Learning Tools.

Step 1

Download Virtual Box

Select the version for your OS e.g. Windows or MacOS

Step 2

Download Ubuntu 16.04

You can download Ubuntu 18.04, however performance on Virtualbox is not as good as 16.04

Step 3

Download Anaconda with Python 3.7+

After downloading Anaconda, go to the folder where it was saved (assuming it’s downloads) and then bash the .sh file to install.

cd Downloads
bash Anaconda3-5.3.0-Linux-x86_64.sh

Follow install instructions (agree to Terms and Conditions and keep pressing “yes” to agree).

Step 4A

Launch Anaconda Navigator is installed launch it by entering the following in terminal.

You can find your Terminal here:

Step 4B

Enter the following line in your terminal window and then hit the Enter key.

anaconda-navigator

The Anaconda Navigator will load.

Step 4C

Go to the Environments tab on the left (see red box above), and then click Create environment (see red box above).

Create a new environment called “cv” (or anything you like) using Python 3.6 (not 3.7, as of Nov-2018 TensorFlow was not supported in Python 3.7+).

Step 4D

Go back to your terminal.

Activate your new environment by typing the following in Terminal.

source activate cv

Assuming ‘cv’ is the name you chose for your environment.

pip install --upgrade pip
pip install tensorflow
pip install keras
conda install scikit-learn
pip install PIL
conda install -c conda-forge opencv
python -m pip install -U matplotlib
pip install pandas
pip install tqdm
pip install pydot
pip install scikit-image
pip install graphviz

If the above line for graphviz doesn’t work go to this link below:

https://packages.ubuntu.com/search?keywords=graphviz&searchon=names

Download and install the version for xenial (16.04LTS)

Step 5

Go to Anaconda Navigator and change environment to the one you just created and and install jupyter notebook.

You’re done!