INSTALL.txt

Python 3 is required for this lab.

If you have problems with these installations instructions, visit tensorflow.org for
additional installation options (virtualenv, docker) and follow the Python 3 instructions.

MacOS:
	If you do not have it already, install git form https://git-scm.com/download/mac
	Install the latest version of python 3 from https://www.python.org/downloads/
	pip3 install --upgrade https://storage.googleapis.com/tensorflow/mac/tensorflow-0.8.0-py3-none-any.whl
	pip3 install --upgrade matplotlib

Windows:
    (This code lab should work under cygwinX, however it had not been tested, which is why Virtualbox is recommended)
	Install Virtualbox from https://www.virtualbox.org/wiki/Downloads
	Create an Linux/Ubuntu virtual machine and install Ubuntu from http://www.ubuntu.com/download/desktop/
	[Optional: In the virtual machine install Virtualbox helpers (allows screen resolution changes) using:
	 sudo -H apt-get install virtualbox-guest-additions-iso
	 sudo -H apt-get install virtualbox-guest-dkms
	 Open “Additional drivers” app and check that the “Virtualbox Graphics Adapter” and “Virtualbox Guest Service” are selected.]
	Shut down the virtual machine, max out its video RAM and CPU settings, allocate 4GB of RAM and restart.
	Now go to Ubuntu/Linux section.

Ubuntu/Linux:
	sudo -H apt-get install git
	sudo -H apt-get install python3
	sudo -H apt-get install python3-matplotlib
	sudo -H apt-get install python3-pip
	curl -O https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.8.0-cp34-cp34m-linux_x86_64.whl
	mv tensorflow-0.8.0-cp34-cp34m-linux_x86_64.whl tensorflow-0.8.0-cp35-cp35m-linux_x86_64.whl
	sudo -H pip3 install --upgrade tensorflow-0.8.0-cp35-cp35m-linux_x86_64.whl
	sudo -H pip3 install --upgrade cycler

    The cycler install is to work around a bug in the current matplotlib release. It should be fixed in the next one.
    The renaming of the tensorflow install file is to overcome a bug in tensorflow. This should get fixed too.

TEST YOUR INSTALLATION:
    git clone https://github.com/martin-gorner/tensorflow-mnist-tutorial.git
    cd tensorflow-mnist-tutorial
    python3 mnist_1.0_softmax.py
    => A window should appear displaying graphical visualisation and you should also see training data in the terminal.