Get Your hands Dirty in these before approaching Machine Learning field
As I always have a keen interest in computers, it made me think of other possibilities we could have. Not just if-else code far more things, like could it predict words based on previous data or analyze a pattern. So I came across a name Machine learning.
As there are many other blogs post out there telling you how to get started, they just tell you a sip of how to work with a library like Pandas, Matplotlib, Seaborn, and many more. It’s great you don’t need to have complete knowledge about Pandas functionality.
After all, you must know how to read the documentation and finds things according to your needs.
Let’s dive in
As the most preferred language in the field of ML is Python, but you’re free to choose any, but I prefer Python. Community support is much better. If you know at least one language, it will be cherry on a cake to dive into Python. You will love the simplicity of it; after that, there’s no coming back.
2. Know how to use Git commands and terminal
Learning terminal & git will be beneficial for you, even if you’re not applying for an ML field. You will get to know about the power of the command line
In coming machine learning concepts, you need to apply and draw meaning from data using Statistics. You will use probability in Naive Bayes and many more areas too. It’s better to get your hands dirty on this.
4. Mathematics related to Machine learning
When you proceed in-depth, you will know everything is mathematics. Somehow like in Regression where we work with Linear & Polynomial Regression, which uses linear equation methods, and it’s aspects. I’m just sharing some snippets from topics to understand it’s necessary to have basics mathematics knowledge, e.g., Linear Algebra
You better know how to manipulate the data using pandas according to need. Like you want to arrange the data using Year and Month in the dataset, you better understand how to tackle that problem. This can be done using Groupby function in Pandas. So its good to have a good grip on Pandas which will be quite powerful for you in future
As we know, pictures say a lot about the problem than just words, so manipulating with data can be visualized using Matplotlib, seaborn, and other libraries. It’s essential if you want to proceed in data science too. In data science, we don’t visualize using aggregate data, but instead, we use a subset of a dataset and draw a conclusion from it.
You better know it’s not about writing code and drawing plot for visualization. It’s all about drawing findings from those plots with high accuracy. Also, you must know there is more inference we can draw from a dataset based on the different plots; it changes sometimes depending on the context. You might draw a dangerous wrong conclusion from it. You should cover all aspects to look at a dataset.
Statistical reasoning is the way people reason with statistical ideas and make sense of statistical information. Statistical reasoning may involve connecting one concept to another (e.g., center and spread) or may combine ideas about data and chance. Source
All the above knowledge is crucial to understand better any problem that you want to solve using machine learning
9. Bingo !!
you’re ready to Enter the world of Machine learning
You can cover all these topics in just one place: Datacamp, which is an excellent platform for all those beginners and intermediate people who want to learn by doing it.
Some resources for you to start
- For Linear Algebra, Multivariate Calculus, Principal Component Analysis
- For Equations, Functions, Graphs, Differentiation, Optimization, Vectors, Matrices, Statistics, Probability
- For Statistics, Probability
- For Probability, Bayes Theorem, Statistical inference, Priors and Models for Discrete Data, Models for Continuous Data
You can reach me out at any time, and I’m happy to help you.
Follow me on Linkedin, Github.
Criticism and Suggestions are welcome !!!
I’m happy to hear from you
