Can you write a software code that automatically learns from data?
Do you believe that computer programs have already started to see, hear and learn?
Do you think that when you go from 1 million training examples to 10 billion training examples for model, it all starts to work?
Do you agree that machine learning and deep learning will create a new set of hot jobs in the coming years?
If you say yes in response to most of the question above then you are already on boarded to machine learning technology.
What is Python?
Python is an interpreted, general-purpose programming language. It was created by Guido van Rossum in 1991.
What is Machine Learning?
Machine learning is a subset of artificial intelligence, where a software system learns from data automatically.
Lets understand the Machine learning model training. It is one of the most important step in machine learning project life cycle. At high level, First we analyze the data and problem statement, and then select a Machine Learning Algorithm.
Then we go for model training and once the model is trained we use the test data to predict the outcome.
The difference between actual value and predicted value is called the loss and our aim is to minimize the loss to maximize the model performance.
Machine learning workflow:
Machine Learning is not about model training only. There are other important steps are involved in Machine Learning workflow.
The high level steps involved are namely, Importing the data , Pre-Processing the data , Doing Exploratory Data Analysis, then of course training the model and finally evaluating the model performance.
Why Python for machine learning:
Python is a simple language and getting up to speed for a developer is quick if he or she has used any other language having C like syntax.
Python has great community, which facilitates good documentation and useful answers in websites like Stack Overflow.
There are many useful libraries for Python, which solve many problems and machine learning is one of them.
Must know libraries of Python:
If you starting out and you want to cover everything from pre-processing to training and testing the machine learning model then go for scikit-learn.
At present, one of the most famous deep learning library in the world is undoubtedly Google’s TensorFlow.
Keras is a wonderful library that provides a high-level API for neural networks and is capable of running on top of TensorFlow.
Facebook created a Python implementation of Torch library, that is called PyTorch in 2016. PyTorch has got some really cool features for deep learning.
Therefore, Machine Learning is the technology of future and Python is the most suitable language for Machine Learning at the moment.
You may like to go through the video tutorial below, which is an extension of this post.
Happy Machine Learning :)
