If you don't like to read, You haven't found the right way !

Web Street Smart

How To Learnt Machine Learning

Spread the love

hi everyone, in this article i show you how i would learn machine learning if i could start over for context i’m a machine learning developer advocate at assembly ai and before that i worked several years as software developer and ml engineer and i also teach python and machine learning on my own   course channel so i would say i’m pretty experienced in the field but i know that the available courses out there can be overwhelming for beginners so i hope to give you some guidance with this article the demand for machine learning engineers is still increasing every year so it’s a great skill to have i divided this learning path into seven steps that should take you about three months to finish of course this can vary depending on how much time and effort you want to put into this and i know that everyone learns differently or might have different goals so this is just my personal take on how to learn machine learning you can use this guide if you just want to explore machine learning as a hobby but also if you plan to find a job in the field i will mention a few more tips about the job search in the end so let’s jump into the study plan.

 the first thing i recommend is to lay the foundation with some math basics now you might say math is not really necessary anymore and this is partly true the available machine learning frameworks abstract the math away and i know many machine learning engineers so don’t need it in their day job at all however in my opinion knowing the underlying math provides you with a better foundation and better understanding of how the algorithms work and it makes your life easier when you run into problems also i think there is beauty in the underlying math that makes the machines learn so for me knowing the math sparked my excitement even more now you don’t need to get too deep into this a great website with free resources is khan academy so my recommendation is just to take some basic courses and then move on and then later when you do the actual machine learning course and don’t understand everything then come back here and learn the missing topics oh and by the way you find all the resources and recommended courses in the othere articles below .

the next step is to learn python it is the number one programming language for machine learning and there is no way around it all major machine learning frameworks are built with it and all major courses use python for their exercises so having decent python skills is essential to build machine learning projects now you don’t need to become an advanced software developer but a little bit more than the beginner level would be great one great thing about python is that it is very beginner friendly and in my opinion it’s the best first programming language you can learn i recommend two free courses on   course one four hour beginner course and one six hour intermediate course and then you should have a solid base this step is what i call the machine learning tech stack and consists of the most important python libraries for machine learning data science and data visualization this step is a bit optional because you can also pick up these skills later when you do the actual machine learning course but i think it’s great to build the foundation first and then it will be easier later the three libraries i recommend at this point are numpy which is the base for everything pandas which is important for data handling and matplotlib which is needed for visualization these libraries are used in almost every machine learning project that’s why i would include them in your learning path at this point again you don’t need to learn too much here i recommend just following one free crash course for each library and then later pick up more advanced concepts if you need them at this point you don’t have to learn the machine learning courses like scikit-learn tensorflow or pytorch you could of course if you want but these are included in the machine learning course i show you in a moment so you can pick this up later now that we’ve covered the coding skills it is finally time for the actual machine learning course there are many great ones available but the most popular and in my opinion also one of the best ones is the machine learning specialization by andrew ng on coursera this specialization includes three courses it got revamped just a few months ago and now includes python with numpy scikit-learn and tensorflow for the code so you not only learn all the essential machine learning concepts but also get your first hands-on experience with the ml libraries it is extensive and it takes several weeks to finish but it’s worth it after these courses i have one more recommendation for you i suggest to implement a few algorithms from scratch in python using only pure python and numpy for example by following my ml from scratch playlist here on   course this is completely optional but it helped me to properly understand some of the concepts from andrew’s course and a lot of students have told me the same feedback so check it out if you want to also we plan to release an updated version of the ml from scratch course here on the assembly ai channel so make sure to subscribe to our channel and don’t miss it now i recommend getting even more hands-on and learning more about data preparation for this kaggle has awesome free courses on their website i recommend at least the intro to ml and intermediate ml courses they are lightweight compared to the previous one and some material is just a refresher for you but you learn more about data pre-processing and data preparation with pandas each lesson has a theory part and then some coding exercises it also gives you a gentle introduction to the kaggle platform and you learn how to make code submissions on kekkel which is perfect.

for the next point now it’s time to practice as much as possible and apply your knowledge to real world machine learning problems for this the best platform is kaggle.com it provides thousands of different data sets and challenges where you can participate participating in challenges can motivate you a lot now i wouldn’t try making it to the top or even winning price money with this because to be honest this requires true expertise and also a lot of gpu power but i would still try to tweak your solutions multiple times by learning more about data pre-processing and also about hyper parameter tuning you can then use kaggle competitions to build your portfolio and put them on your cv so in my opinion keggle is an awesome platform and you should practice here as much as possible of course you can also tackle other machine learning problems outside of kaggle it just makes your life a little bit easier because it provides you with the data sets a platform to evaluate the projects and there’s a whole community around it at this point you can already be super proud of yourself and now in this last section i want to give you a few more tips if your goal is to get a job the tasks of ml engineers vary a lot and it’s not possible to know everything for example some positions are specialized in computer vision or nlp or they require you to have experience with a specific ml framework or even ml ops requirements like how to deploy and scale ml apps mlops is a whole field on its own so i may cover this in a separate article my point is you have to decide in which field you want to work and then look at the requirements and some corresponding job othere articless and then specialize in this direction another great tip i can give you and this is something i wish i had done earlier in my career is to start a blog you can write tutorials share what you’ve learned which projects you’ve built which problems you’ve faced along the way and how you’ve solved them by writing about a topic you can deepen your knowledge and then you can use this as a resource on your cv trust me this will increase your chances to get an interview a lot alright that’s my recommendation for a machine learning study guide again this might not be suited for everyone this is just how i would learn machine learning if i had to start over just one more quick addition if you prefer learning with books then you can check out these two books let me know in the comments if this was helpful or if you have any other suggestions you would add to the plan don’t forget to check out the resource list i put in the othere articles below and then i hope to see you in the next article .

Similar Articles

Comments

LEAVE A REPLY

Please enter your comment!
Please enter your name here

Instagram

Most Popular