A Sneak Peek Into ML
We often have viewed the recommendation list crafted especially for us, while binge-watching on any of the OTT platforms such as Netflix, Amazon Prime, etc. Have you ever thought who could be the mastermind behind theses??
E-commerce companies and content providers use recommender systems to suggest related products or content to users. Even there is still plenty of room for improvement, recommender systems are one of the most successful stories in machine learning ML.
The reason I selected this topic as my first article to write is because somewhere down the line I feel the pursuits needed for a beginner to dive in learning ml are just missing. Sometimes it’s just the missing or incomplete info that is required for one to start exploring new technology.
In this article, I will try to cover all the topics for the ones who are taking baby steps in learning ML!
What is ML?
Machine learning is an application of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed [1].
Talking out of definitions and technical terms, we can state machine learning as training a model (model can be a machine) by applying some algorithms (the one we write in any languages) and then testing that model by giving a different set of data
For eg. I fed 20 photos of banana and 20 images of mango to a machine. The machine now learns to classify any image as mango or banana on the basis of provided input. Now, if I show a new image that is unfamiliar to the machine; the machine now correctly identifies whether the given image is of banana or mango.
EXPLORATION PROCESS:
If you are someone who wants to learn ML but are ignorant of methods and resources, something is below for you!
To learn anything new or come into the expertise of any domain, one needs to follow some particular steps to achieve it, else one can get stuck in an infinite loop of knowing and researching and not getting ahead in any fields.
Over my experience of learning and teaching ML, I have observed and identified a pattern to learn ML. Here below
are some steps which you need to follow:
1. Learn any programming language you seem is awesome! (Always remember, the language you choose should be independent of public opinion and fame statistics)
2. Know the process of ML, what are its types and the entire Performa about its functioning.
3. Start learning libraries required for training a model
I have listed below some of the Python libraries which will be your building bricks in learning ML
NumPy:
For computing mathematical calculations and creating data structures, etc.
Pandas:
Reading Data and Processing it for fitting on our model
Matplotlib:
Library for plotting equations or determining co-relation between features of your data
Seaborn:
Visualization Library used for plotting in data
Scikit-Learn:
Used for creating ML algorithms
Tkinter:
Used for creating GUI in Python
Plotly:
Interactive Visualizations are created using this library
Beautiful Soup:
Navigating webpages and using this can perform different web applications part.
And many more….
And the list continues to infinity….
Don’t worry about watching such a big list; neither of them is difficult to study and implement; but, are the easiest ones that we can use!
You might be worried about the time taken to study and your capability.
But always remember “NO PAINS NO GAINS”. So, the proverb wins and if you are self-determined and you know what you want to do! Nobody’s going to stop you from getting it!
HAHA! Too much philosophical.
Moving forward, once you have learned these libraries, you are now ready to build ml models and train data and see the results!!
WHY Machine Learning?
You might be thinking so much to learn, such a long process. Isn’t it painless to just learn to code and being hired as a software developer or just a coder!
The answer is Sure, it’s excellent being in that place, but I always recommend my peers to go beyond imagination. If you are learning code then y just code and why not apply your coding knowledge to create something new! Which is the motive of being an engineer right?
Practically speaking, besides is a statistic why one should adapt ml and how it can help boost a career if one stays committed and never stop the cycle of learning and implementing.
AGREED TO LEARN, BUT FROM WHERE?
Some of you might have stopped reading this at the very beginning, some might have thought ML isn’t my forte and some might have been thinking to just explore this one.
For all of you! What I feel is everyone must once just dive into this domain and take an essence of it! And then decide whether to continue or just switch into another one?
Here is a list of resources one can refer to learn ML.
1. Python beginner course from udemy:
It’s a must course to learn python if you are a beginner and have no prior experience with python language or coding.
· https://www.udemy.com/share/101W8Q/
2. Python for Data Science and Machine Learning Course from udemy:
This course is just an ideal course to teach ml and it’s subsequent libraries. I highly recommend this course!! Do watch it!
· https://www.udemy.com/share/101WaU/
3. Python language course from Coursera:
This is another beautiful course from Coursera which teaches you A to Z of Python for new ones to these fields!
· https://www.coursera.org/specializations/python
CREDITS:
· Freepik and Google for providing me such great graphics!
