First Step to Machine Learning

What is Machine Learning???

Machine learning (ML) is the scientific study of algorithms and statistical models that computer systems use to effectively perform a specific task without using explicit instructions, relying on models and inference instead. It is seen as a subset of artificial intelligence. Machine learning algorithms build a mathematical model of sample data, known as "training data", in order to make predictions or decisions without being explicitly programmed to perform the task. (Wikipedia)

Machine Learning Approaches

Getting Start

We hope to do these tutorials using TensorFlow  (TF) machine learning library which is a open source product of Alphabet.

TF is supporting for following Operating Systems,

• Ubuntu 16.04 or later
• Windows 7 or later
• mac-OS 10.12.6 or later
• Raspbian 9.0 or later

Here I'll show the installation in Ubuntu. If you are using different OS please use this link and do the installation.

First Step:

Install Python

Open a terminal (Ctrl + alt + t) and paste following command

sudo add-apt-repository ppa:jonathonf/python-3.6

(Here I'm using python 3+ as I'm not well familiarized with Python 2.+. It is OK if you can manage python 2.+.)

Then paste the following command.

sudo apt-get update

sudo apt-get install python3.6

 

OK. Python installation is finished now. To verify that you've got Python type following command.

python3 -V

This will display your python version.

If you are using python 2.+ you don't need to install it as Ubuntu already has it.

Install pip

So the next step to install pip. (We use pip to install TF in our machine). Nothing hard to do, just paste following command in terminal.

sudo apt install python3-pip

If it is not working use following one,

sudo apt-get install python3-pip

So pip installation is done.

Install tensorflow (TF)

Just paste the following command in terminal and wait. pip will do the rest.

pip install --upgrade tensorflow 

If the above command didn't work use following,

pip3 install --upgrade tensorflow 

(This is the current stable release for CPU only. My advice is to use this. If you want to use GPU for training process you can find those installation methods in here)

So it's done. To check whether tf installed or not, paste following command in terminal.

python -c "import tensorflow as tf; tf.enable_eager_execution(); print(tf.reduce_sum(tf.random_normal([1000, 1000])))"

In the next article we'll see how to build a tf model and train it.