Kick-starting your ML journey

Have you ever wondered , sitting in a monotonous algebra class — “Where am I gonna actually utilize this boatload of algebraic nonsense ever in my life, I mean who really wants to find “x” ?”

Well, I FINALLY happen to have an answer !

Machine learning is a perfect example of the implementation of Linear algebra , Statistics and Calculus.

AI is a multidisciplinary field made at the convergence of, and by the collaboration between, software engineering, measurements, neurobiology, and control hypothesis. Its rise has assumed a key job in a few fields and has on a very basic level changed the vision of programming.

If the question before was, how to program a computer,now the question becomes how computers will program themselves. Thus, it is evident that machine learning is a method that allows a computer to have its own intelligence.

But when it comes to the application of Machine learning people tend to skip the hard part, Mathematics again and jump on to the hands on part right away!

So, Why is Mathematics important?

1. Selecting the right algorithm which includes giving considerations to accuracy, training time, model complexity, number of parameters and number of features.

2. Choosing parameter settings and validation strategies.

3. Identifying underfitting and overfitting.

4. Adjusting the model to get ideal accuracy.

5. Estimating the correct confidence level and uncertainty.

The fact is maths, at times, does get difficult, and there WILL be a few courses that skim through it to give you a ‘hands-on’ experience on the topic, but deep down, when it comes to it, machine learning is more of a mathematical subject than a computer science one.

Without the maths you can figure out how to write the code, but that’s like figuring out how to write out sentences in a language without understanding the grammar.

You can do it, but it’s as good as a guess what the outcome is gonna be!

Bottom line, it is possible to be a functional data scientist without being a mathematical wizard. The reality is that, in the industry, data scientists spend a huge amount of their time in data cleaning, data exploration, and data gathering. In all actuality 80% of them are occupied with information perception and information fighting.

Getting started with programming:

Beginning with programming: For fledglings, writing computer programs is in some cases terrifying to learn. Normally, it may be scary and troublesome from the start. In any case, with standard practice, this ability can be aced in the long run. Coming to programming dialects in ML, Python stands apart top because of its adaptability and straightforwardness in coding. Different dialects, for example, R, Julia and Java additionally help in building ML ventures.

● Start with a basics course on Python Programming.

● Start with the basic libraries like Numpy, Pandas, Matplotlib, Sklearn.

● Compute algorithms from scratch to build a foundation.

● Practice on various datasets to test your knowledge.

Resources :

1. Learn hands-on implementation on Datacamp or courses available on Coursera, Udemy, Udacity, etc.

2. Lookup documentations.

3. Practice on Kaggle.

Applying machine learning:

1. Get Data (webscraping, API calls, image libraries): coding background.

2. Clean/munge the information. This fair takes some messing with the coding foundation.

3. Tune your algorithm. Adjust all your bells and whistles on your model to try to get a good overall model (measured with accuracy, recall, precision, f1 score, etc — you should look these up). Then check for overfitting/underfitting etc with cross-validation methods (again, look this one up): math background.

4. Visualize! Here’s where your coding background pays off some more, because you now know how to make plots and what plot functions can do what this gives you an idea of data science.

5. Expand your knowledge on different algorithms and its implementation.

Once you delve deeper into machine learning you find out that there are a lot of fields to explore and the more you read and learn, the more interesting it becomes!

Happy learning 😃