(Note to Self - How I would learn Machine Learning) 01:00 1. Math: Khan Academy Recommended Courses: - Multi-Variable Calculus - Differential Equations - Linear Algebra - Statistics and Probability 02:00 2. Python Recommended Courses - FreeCodeCamp: Python in 4-Hours Full Course - FreeCodeCamp: Intermediate Python in 6-Hours 02:37 3. Machine Learning TECH STACK Most important Python libraries for Machine Learning, Data Science, and Data Visualization Optional: Can be picked up later when doing the ML course. Use for every project, which is why he recommends doing them now to build a base. Follow a free crash course for now, pick up more advanced concepts later if needed. - NumPy: Base for everything: Python Engineer - NumPy Crash Course Complete Tutorial - Pandas: Data handling: Keith Gali - Complete Python Pandas Data Science Tutorial - MatPlotLib: Visualization: FreeCodeCamp - MatPlotLib Crash Course --------------------------- The following MachineLearning courses aren't yet needed - Tensor Flow - Scikit Learn - PyCharge ??? 03:35 4. Machine Learning Courses - Machine Learning Specialization by Andrew Ng (Coursera) - Implement algorithm from scratch using his 'ML from SCRATCH' playlist - ML from Scratch Playlist by Python Engineer (Assembly AI) 04:45 5. Hands - On & Data Preparation Kaggle Courses - Intro to Machine Learning - Intermediate Machine Learning 05:19 6. Practice & Build Portfolio Kaggle: Competitions - They provide lots of datasets, platform to evaluate, and a community. 06:15 7. Specialize & Create Blog - NLP - PyTorch / Tensor Flow - MLOps 06:52 Start a VLOG - Tutorial - Share what you've learned - Share the projects you've built - Problems faced and how you have solved them - Write about a topic 07:24 Books - Machine Learning with PyTorch and SckiKit-Learn by Raschka - Hands-On Machine Learning with SciKit-Learn, Keras & TensorFlow by Geron
Trying out this roadmap March 1st 2023. Will update everyone 6months from then. I’m already a software engineer so I’ll be skipping the coding steps and the math will be refreshers but far from a data scientist or data analyst for that matter. Hope everything works out. See you guys in the future!
I really value this plan...you don't understand. There's so many people who quit at the jump because people in the industry give very broad steps. This is a very clear plan with flexibility to go even deeper into each resource and step. Also, for starters, you even said 3 months. Some may say that is unrealistic but as a Math major with no CS experience but a heavy interest in AI theoretically, the drive is already there. Learning can't be rushed but it can definitely be integrated quickly with the right resources. I plan on putting at least 10 hours each week into this journey. Thanks again man!
@@ameynarwadkar7924 well, since OP is MIA, I'll give you my update. For context, I have a bachelor's in electromechanical engineering, so I skipped the math courses for now. I also have a ton of experience using MatLab, so I already have a solid fundamentals on coding logic, objects, and loops. Since I left my comment, I've gotten through the beginner python courses, half of the intermediate Python course, and I'm starting on the ML Tech Stack this weekend. The beginner python course was very helpful. He goes through some of the Python fundamentals by coming up with real-world problems, and then using the concepts he shows you to solve those problems. A word of advice: after he explains what he's about to do at the beginning of each tutorial, pause the video, and see if you can do it yourself. Be persistent. Then play the video, and compare what you built to what he shows you. It will take you much longer to get through the video this way, but I think it's a much more effective way to learn for most people. After the beginner course, I refined some of the basics by building my own simple programs of things I came up with. Example: I built a program that calculates a list of prime numbers in a user defined range, I wrote a script that approximates pi using a random number generator, I wrote an algorithm that calculates the largest number in a list of randomly generated integers.... Stuff like that. Simple logic puzzles that will help build your confidence and refine some of the basics in a practical context. I got about halfway through the "intermediate" video and realized it was kind of a waste of time for me. He doesn't actually discuss any intermediate concepts. He just lists off a whole bunch of miscellaneous functions that you may or may not use. He doesn't discuss where the functions would be used, or demonstrate how to solve a problem using the functions.... He just explains the function's syntax, and moves on. And frankly, I'm not going to remember 90% of it anyway, so I decided to skip the rest. I figure if there's a new function I need to use in the future, I'll just Google the syntax and proper use when I need it. But that's just me, and how I learn. If you're one of those people who have a photographic memory, or you plan on making syntax flashcards or something, then maybe this video will be useful to you. But personally I don't learn that way. The "ML Tech Stack" I'm just starting now, so I can't really speak to that yet. I plan on breezing through that pretty quickly. And I can give you another update once I start the actual ML stuff.
This is just what I was looking for! I was overwhelmed with the amount of resources out there, so it is incredibly useful to have a solid roadmap going forward. Thank you!
Heads up regarding the math course recommendations - you can't just do things like Multivariable Calculus out of the blue without proper background. That's the equivalent of Calculus 3 at my school, so I recommend completing Calculus 1 and 2 before knocking out the Multivariable course or any of the others for that matter - best of luck knocking out the course requirements!
Hi Sir, Thanks so much for this roadmap. I want to learn AI & Machine Learning but I had no idea on how to do so. But your video explains everything I need to do. Thanks so much.
Amazing how machine learning algorithm in UA-cam works, I was just thinking about ML earlier today and in the evening I got this video recommendation. ☀🔨
I liked this video and saved it. Made me also notice something about IT people: they don't breathe! I was listening to an IT specialist on TV yesterday, he didn't even listen to the Qs of the interviewer my head hurts its even unsettling
Thanks for the video. I have learned lots of ML-related stuff in the past several months, but I feel like the way I have learned is NOT the the best way. The way you suggested makes more sense.
Thanks for the great learning plan. I would just add that for Multivariable Calculus, Single variable calculus is needed. And as an option instead of "Statistics Probability" i would use an ordered learning path: "Combinatorics -> Probability -> Statistics"
@@adekanbioluwaseun219 Hi. Not yet. I would say i've just start to learn the Math and Python. I am not sure i will became a ML engineer, but along the journey i will definitively pick up a lot of skills.
My leaning map is below,hope it can help anybody i begin my python learning in 12,2022. I quickly read a book in 72h. After that I began to learn on Kaggle. I make my coding skills better(some basic pandas numpy and matplotlib) and learn some basic ml. After that I find that my data analysis skill is not good enough to clean the data. I read the book called python for data analysis. I finished it in 2,2023. From 3,2023 to now I am reading the ml part of hands on ml. I hope to finish it in next week. This book give us some real world views. After that it’s time to learn the math inside the ml.
I think the most underrated part is the math. I myself study Artificial Intelligence in university, which is a bit different and more advanced than simply machine learning. We take 12 courses upfront before starting 'the real deal' machine learning. We learn linear algebra, calculus, bayesian statistics, logic and I absolutely love the way our major is structured in this way, because now that we're doing machine learning, everything makes sense and with this knowledge you really learn on what data you can apply which model. You don't learn that online. They simply say: "for these problems, you simply use these models", which is okay for data scientists, but not for people who study AI themselves within the research field.
Great video! I was completely lost on how to start learning about AI. I am a finance major, and I realized that if I don't learn it now, I will probably get left behind.
I've been learning AI for 20 years and I still have much to learn - IMHO, doubltful that someone could grok a critical mass of these concepts in 3 months. There is a world of difference from being able to ETL a dataset or fine tune a commodotized HF pipeline + tokenizer, and engineer a novel model architecture.
Any science, engineering, technology field will require a foundation of strong math skills, so it's VERY important to brush up on these skills or learn them in a university setting before moving on to the next steps of programming in python. Then you'll learn about python data structures and apply what you've learned to implementing machine learning algorithms. And so on!
Anyone who says you don't need to understand calculus or matrix algebra to learn basic machine learning algorithms and methods (LASSO, Ridge, sample splitting, k-fold cross validation, Stepwise Regression, Random Forest, K-Nearest Neighbors) is technically correct it is possible. Anyone who says this about advanced methods like neural networks and deep neural networks (deep learning) is also correct, but there's a catch, the explanations of what a neural network is, and why/how it works you can find on the internet which can be understood by someone without calculus or matrix algebra are RIDICULOUSLY tedious and painful and overly complicated. Once you already understand that math and what standard/classical multiple (linear) regression analysis in statistics and econometrics is before learning machine learning as I did. What a neural network is just becomes something you can understand within a paragraph or two lol
Thanks for this. I’ve got a physics ms, so I have a lot of the math, and some basic python(I enjoy Python but really only had time to learn it while using it for physics stuff), and I’ve been looking at where to go next to better understand machine learning.
Hello guys, I am ready to become an ML engineer, I'm going to follow this path, and I'll be updating my progress, f*ck motivation, this is about habits. Let's go.
I like the general approach given here, and I am familiar with many of the learning resources and recommend them. However, I do not think it is reasonable to learn all of the material listed here in seven months. That would be rushing things, to put it mildly. With each topic--math, general programming, ML stacks, ML--you would learn to complete exercises, but not how to address new sitatuions. It takes longer than seven months, in my opinion, even if you devote yourself 100%.
Ok, now I'm realising I just have to remember the math stuff. I'll give each course (multivariable calculus, differential equations, linear algebra and statistics probability) a week. In a month from now, I should have a strong math foundations
A week went by and I'm just starting with multivariable calculus (life has been hard). I said that I would give each math course one week, but man I love math. I will finish every course instead of just learning the basics because something tells me that I will need that information later when I finish the machine learning journey
Now that the second week has passed, I can now say that I'm concerned about the time I have spent studying. I have multivariable calculus on pause because I have to study linear algebra to understand every concept needed. Also, I didn't mention this, but I'm supposed to build an AI for a bet I did with a friend. I might jump straight to coding before finishing math to speed up things. See ya next week!
Third week completed and guess what, I'm still with maths. I've got a new hobbie (jugger) and you could say that I sure am procrastinating, statement that is true BUT progress is being made. Will I be able to learn the coding part in one month so I can study math for 2 months now? Probably not, but it's worth the risk
A month has been completed. Linear algebra is about to be finished and I'll finish multivariable calculus in a week or two. I can't wait to start coding :( Differential equations and statistics probability haven't been touched yet, but their time will come, I guess...
Hi Patrick, I'm concerned about ( Machine Learning Courses 3:35 ) the specialization is not free unless you take courses one by one and audit them. Is that what you meant!!! Auditing the course... since auditing means I can't participate in the homework and projects??? What is your recommendation for either auditing the course and practicing from your playlist (Machine Learning from Scratch - Python Tutorials) or enrolling and paying the tuition in Coursera to fully utilize the course tools, such as projects? Thanks
numpy is exactly what you need for mathematics. There is a book on linear algebra and Python. Not take book for Theoretical Linear Alghebra (it not for programmers)
Copy paste of another comment. will update my progress here. (Note to Self - How I would learn Machine Learning) 01:00 1. Math: Khan Academy Recommended Courses: - Multi-Variable Calculus - Differential Equations - Linear Algebra - Statistics and Probability 02:00 2. Python Recommended Courses - FreeCodeCamp: Python in 4-Hours Full Course - FreeCodeCamp: Intermediate Python in 6-Hours 02:37 3. Machine Learning TECH STACK Most important Python libraries for Machine Learning, Data Science, and Data Visualization Optional: Can be picked up later when doing the ML course. Use for every project, which is why he recommends doing them now to build a base. Follow a free crash course for now, pick up more advanced concepts later if needed. - NumPy: Base for everything: Python Engineer - NumPy Crash Course Complete Tutorial - Pandas: Data handling: Keith Gali - Complete Python Pandas Data Science Tutorial - MatPlotLib: Visualization: FreeCodeCamp - MatPlotLib Crash Course --------------------------- The following MachineLearning courses aren't yet needed - Tensor Flow - Scikit Learn - PyCharge ??? 03:35 4. Machine Learning Courses - Machine Learning Specialization by Andrew Ng (Coursera) - Implement algorithm from scratch using his 'ML from SCRATCH' playlist - ML from Scratch Playlist by Python Engineer (Assembly AI) 04:45 5. Hands - On & Data Preparation Kaggle Courses - Intro to Machine Learning - Intermediate Machine Learning 05:19 6. Practice & Build Portfolio Kaggle: Competitions - They provide lots of datasets, platform to evaluate, and a community. 06:15 7. Specialize & Create Blog - NLP - PyTorch / Tensor Flow - MLOps 06:52 Start a VLOG - Tutorial - Share what you've learned - Share the projects you've built - Problems faced and how you have solved them - Write about a topic 07:24 Books - Machine Learning with PyTorch and SckiKit-Learn by Raschka - Hands-On Machine Learning with SciKit-Learn, Keras & TensorFlow by Geron
Math. Thank fudging god. So many of the well known tech UA-cam data scientists cover the same topic, unfortunately they are just glorified business analysts (no disrespect to BAs or DAs their profound knowledge in domain and data is what drives the industry). However, a true data scientist is a statistician/mathematician first. Programming is literally a language. I cannot be more literal when I say language, just as English , Latin, Sanskrit are languages to communicate, programming is simililar. Python has a lot of “vocabulary” (libraries),accommodating machine learning, we can implement any ml algo using any language. That’s it. If you have a proper mathematician and a proper coder that’s the ultimate combo. Design the algo code the basics, give to a proper coder and they will optimize it the compute. Cheers mate. Subbed.
Why do you want to move away from Full Stack Engineering? I'm a fresher who's being trained in full stack, and I'm wondering if it's the best way to go for me. Can you tell me what has turned you away from Full Stack?
Question: For the sake of having a mentor / tutor, do you recommend taking a pricey course on, say Interview Kickstart? They have an $11,000 full AI course that has tutors who work in the industry there to help you at least 3 days a week. For the sake of organization and knowing for sure you're learning the right stuff, I can see that being a good thing, but can this stuff still be learned well enough for much less money to get a job in the industry? Would love you know your thoughts and anyone else's reading this - thanks! In this day and age there are way to many options and opinions out there that don't make sense - it was nice to get a clear and concise short take on what to do. Even though I am a senior in college and have a decent amount of personal experience with programming and full stack web dev and some industry experience with those, I'll be following all these steps to make sure I get the best foundation possible.
(Note to Self - How I would learn Machine Learning)
01:00 1. Math: Khan Academy
Recommended Courses:
- Multi-Variable Calculus
- Differential Equations
- Linear Algebra
- Statistics and Probability
02:00 2. Python
Recommended Courses
- FreeCodeCamp: Python in 4-Hours Full Course
- FreeCodeCamp: Intermediate Python in 6-Hours
02:37 3. Machine Learning TECH STACK
Most important Python libraries for Machine Learning, Data Science, and Data Visualization
Optional: Can be picked up later when doing the ML course.
Use for every project, which is why he recommends doing them now to build a base.
Follow a free crash course for now, pick up more advanced concepts later if needed.
- NumPy: Base for everything: Python Engineer - NumPy Crash Course Complete Tutorial
- Pandas: Data handling: Keith Gali - Complete Python Pandas Data Science Tutorial
- MatPlotLib: Visualization: FreeCodeCamp - MatPlotLib Crash Course
--------------------------- The following MachineLearning courses aren't yet needed
- Tensor Flow
- Scikit Learn
- PyCharge ???
03:35 4. Machine Learning Courses
- Machine Learning Specialization by Andrew Ng (Coursera)
- Implement algorithm from scratch using his 'ML from SCRATCH' playlist
- ML from Scratch Playlist by Python Engineer (Assembly AI)
04:45 5. Hands - On & Data Preparation
Kaggle Courses
- Intro to Machine Learning
- Intermediate Machine Learning
05:19 6. Practice & Build Portfolio
Kaggle: Competitions
- They provide lots of datasets, platform to evaluate, and a community.
06:15 7. Specialize & Create Blog
- NLP
- PyTorch / Tensor Flow
- MLOps
06:52 Start a VLOG
- Tutorial
- Share what you've learned
- Share the projects you've built
- Problems faced and how you have solved them
- Write about a topic
07:24 Books
- Machine Learning with PyTorch and SckiKit-Learn by Raschka
- Hands-On Machine Learning with SciKit-Learn, Keras & TensorFlow by Geron
Gud
Thank you!!
Machine Learners are busy people. Your comment proves you understand that!
Ok
Best comment!
Thank you
1. Math 1:00
2. Python 2:00
3. Machine Learning TECH STACK 2:37
4. Machine Learning Courses 3:35
5. Hands - On & Data Preparation 4:45
6. Practice & Build Portfolio 5:19
7. Specialize & Create Blog 6:15
Awesome! Thank you for sharing.
Thank you. May your thoughtfulness be rewarded a thousand times.
… but the videos has chapters.. 🤨
W pfp
Thank you
learning machine learning is quite fun
starting this roadmap from today. wish me luck!
hope everyone else also achieves their goal.
hello Dude, whats ur progress?
Trying out this roadmap March 1st 2023. Will update everyone 6months from then. I’m already a software engineer so I’ll be skipping the coding steps and the math will be refreshers but far from a data scientist or data analyst for that matter. Hope everything works out. See you guys in the future!
Good luck 🤞 Commenting so I can see the updates
Yessir good luck!!
yes
Leetssssss go!!!
all the best bro, you will make it big, ik it!!
I really value this plan...you don't understand. There's so many people who quit at the jump because people in the industry give very broad steps. This is a very clear plan with flexibility to go even deeper into each resource and step. Also, for starters, you even said 3 months. Some may say that is unrealistic but as a Math major with no CS experience but a heavy interest in AI theoretically, the drive is already there. Learning can't be rushed but it can definitely be integrated quickly with the right resources. I plan on putting at least 10 hours each week into this journey. Thanks again man!
How's it going so far?
How’s your learning journey been ? You must be at the end of it. Give us an update. I am planning on joining the same journey
Give an update bro, I'm also Math major with a lot of interest in AI and on this journey rn.
@@ameynarwadkar7924 well, since OP is MIA, I'll give you my update. For context, I have a bachelor's in electromechanical engineering, so I skipped the math courses for now. I also have a ton of experience using MatLab, so I already have a solid fundamentals on coding logic, objects, and loops.
Since I left my comment, I've gotten through the beginner python courses, half of the intermediate Python course, and I'm starting on the ML Tech Stack this weekend.
The beginner python course was very helpful. He goes through some of the Python fundamentals by coming up with real-world problems, and then using the concepts he shows you to solve those problems. A word of advice: after he explains what he's about to do at the beginning of each tutorial, pause the video, and see if you can do it yourself. Be persistent. Then play the video, and compare what you built to what he shows you. It will take you much longer to get through the video this way, but I think it's a much more effective way to learn for most people.
After the beginner course, I refined some of the basics by building my own simple programs of things I came up with. Example: I built a program that calculates a list of prime numbers in a user defined range, I wrote a script that approximates pi using a random number generator, I wrote an algorithm that calculates the largest number in a list of randomly generated integers.... Stuff like that. Simple logic puzzles that will help build your confidence and refine some of the basics in a practical context.
I got about halfway through the "intermediate" video and realized it was kind of a waste of time for me. He doesn't actually discuss any intermediate concepts. He just lists off a whole bunch of miscellaneous functions that you may or may not use. He doesn't discuss where the functions would be used, or demonstrate how to solve a problem using the functions.... He just explains the function's syntax, and moves on. And frankly, I'm not going to remember 90% of it anyway, so I decided to skip the rest. I figure if there's a new function I need to use in the future, I'll just Google the syntax and proper use when I need it. But that's just me, and how I learn. If you're one of those people who have a photographic memory, or you plan on making syntax flashcards or something, then maybe this video will be useful to you. But personally I don't learn that way.
The "ML Tech Stack" I'm just starting now, so I can't really speak to that yet. I plan on breezing through that pretty quickly. And I can give you another update once I start the actual ML stuff.
Tell us where u at?
Nice, I was struggling to decide what to learn first? This field is so overwhelming for beginners. Thanks for explaining out everything so clearly.
Very welcome!
How did you proceed?
Hi Puneet, how did you learning ML go? I am starting to explore my options in this area.
Hey did you learn any math
Very effective steps! I have been following this roadmap for the past couple of months, and I am happy with the progress I have made
This is just what I was looking for! I was overwhelmed with the amount of resources out there, so it is incredibly useful to have a solid roadmap going forward. Thank you!
all you had to do was to poop and drink some coffee...
Heads up regarding the math course recommendations - you can't just do things like Multivariable Calculus out of the blue without proper background. That's the equivalent of Calculus 3 at my school, so I recommend completing Calculus 1 and 2 before knocking out the Multivariable course or any of the others for that matter - best of luck knocking out the course requirements!
Hi Sir, Thanks so much for this roadmap. I want to learn AI & Machine Learning but I had no idea on how to do so. But your video explains everything I need to do. Thanks so much.
Can we connect? I want to learn also, and a learning partner won’t hurt
@@jamesojih8050 I'm currently busy learning other technologies. So, I wont be able to start on AI at this moment.
Amazing how machine learning algorithm in UA-cam works, I was just thinking about ML earlier today and in the evening I got this video recommendation. ☀🔨
that is mind learning, not machine learning
This man just single handedly planned my life, what a legend!
Agreed my friend
I liked this video and saved it. Made me also notice something about IT people: they don't breathe! I was listening to an IT specialist on TV yesterday, he didn't even listen to the Qs of the interviewer my head hurts its even unsettling
Thanks for the video. I have learned lots of ML-related stuff in the past several months, but I feel like the way I have learned is NOT the the best way. The way you suggested makes more sense.
Thanks for the great learning plan. I would just add that for Multivariable Calculus, Single variable calculus is needed. And as an option instead of "Statistics Probability" i would use an ordered learning path: "Combinatorics -> Probability -> Statistics"
Hi Andrey, are you already an ML Engineer?
@@adekanbioluwaseun219 Hi. Not yet. I would say i've just start to learn the Math and Python. I am not sure i will became a ML engineer, but along the journey i will definitively pick up a lot of skills.
@@andreypopov6166 I think the same but I'm just starting, for now, should i just do the courses patrick mentioned in order?
Hi. How long did it took you guys to complete the math studies
One of the most luxurious pieces of advice I've ever heard ( or watched)
Thank you, Patrick.
Your suggestion to create a blog is simply genius.
My leaning map is below,hope it can help anybody
i begin my python learning in 12,2022. I quickly read a book in 72h. After that I began to learn on Kaggle. I make my coding skills better(some basic pandas numpy and matplotlib) and learn some basic ml.
After that I find that my data analysis skill is not good enough to clean the data. I read the book called python for data analysis. I finished it in 2,2023.
From 3,2023 to now I am reading the ml part of hands on ml. I hope to finish it in next week. This book give us some real world views. After that it’s time to learn the math inside the ml.
The blog tip is great! Gonna use that for sure inshAllah :)))
This outline is phenomenal - thank you!
One of the most luxurious advice I've ever heard ( or watched)
Thank you Patrick
I think the most underrated part is the math. I myself study Artificial Intelligence in university, which is a bit different and more advanced than simply machine learning. We take 12 courses upfront before starting 'the real deal' machine learning. We learn linear algebra, calculus, bayesian statistics, logic and I absolutely love the way our major is structured in this way, because now that we're doing machine learning, everything makes sense and with this knowledge you really learn on what data you can apply which model. You don't learn that online. They simply say: "for these problems, you simply use these models", which is okay for data scientists, but not for people who study AI themselves within the research field.
Can you share the link for the course curicculum or syllabus?
what university do you go to
Intro to Statistical Learning by Gareth James and others is a great book for learning the statistical part for basics.
Thank you Patrik!!! Amazing intro for ML topic🙏🙏
This outline is phenomenal - thank you!. This outline is phenomenal - thank you!.
Great video! I was completely lost on how to start learning about AI. I am a finance major, and I realized that if I don't learn it now, I will probably get left behind.
Also starting with this roadmap today, for now refreshing math. Hope I can get through this 100%!
How is it going bro?
@abdel8819 slowly! But made some progress, started with Andrew Ng courses and trained my first model already;)
Youre awesome, no bullshit, litteraly just helping people, thank you.
This is exact i was looking for. Thank you so much
Awesome content ! Thank you so much for sharing it
I've been learning AI for 20 years and I still have much to learn - IMHO, doubltful that someone could grok a critical mass of these concepts in 3 months. There is a world of difference from being able to ETL a dataset or fine tune a commodotized HF pipeline + tokenizer, and engineer a novel model architecture.
I am starting to follow this roadmap on April 12th. I will update my status every month. Good luck everyone!
Удача тебе, судя по всему, не помогла
Awesome. Go for it. Can’t wait to hear updates.
True and honest roadmap, thanks a lot!!
Thank you for the video, I was lost on how to start
Fantastic. Short, to the point and clear!
Wonderful video! Very useful information was presented :)
thank you :)
Great roadmap Patrick! It would be great if add few examples projects to practise. Most of the ML learners find it challenging to find projects.
Any science, engineering, technology field will require a foundation of strong math skills, so it's VERY important to brush up on these skills or learn them in a university setting before moving on to the next steps of programming in python. Then you'll learn about python data structures and apply what you've learned to implementing machine learning algorithms. And so on!
I come from an excel/SQL/R background and am recently unemployed (by choice). Starting this from math stage beginning July 19, 2023.
Anyone who says you don't need to understand calculus or matrix algebra to learn basic machine learning algorithms and methods (LASSO, Ridge, sample splitting, k-fold cross validation, Stepwise Regression, Random Forest, K-Nearest Neighbors) is technically correct it is possible. Anyone who says this about advanced methods like neural networks and deep neural networks (deep learning) is also correct, but there's a catch, the explanations of what a neural network is, and why/how it works you can find on the internet which can be understood by someone without calculus or matrix algebra are RIDICULOUSLY tedious and painful and overly complicated.
Once you already understand that math and what standard/classical multiple (linear) regression analysis in statistics and econometrics is before learning machine learning as I did. What a neural network is just becomes something you can understand within a paragraph or two lol
Omg perfect for my self made pre-master!!
Nice informative video and helped me build a roadmap. Thanks!!!
Thanks for this. I’ve got a physics ms, so I have a lot of the math, and some basic python(I enjoy Python but really only had time to learn it while using it for physics stuff), and I’ve been looking at where to go next to better understand machine learning.
Very Good Explanation. Great Video. Thank You for you ❤
Thanks for the advice. I’m going to apply your approach in my learning. It sounded feasible and well-though 🙏
Now i know where to learn and in what order TY
Your voice is so f***ing sharp.
Thank you so much. This is extremely valuable.
Glad you think so!
Thank you for making this video. It's very helpful.
Thank you so much Patrick for your insights and guidance. The links you have provided are really helpful. Thanks again!
Hello guys, I am ready to become an ML engineer, I'm going to follow this path, and I'll be updating my progress, f*ck motivation, this is about habits. Let's go.
Me too brother
Thanks man, it’s very helpful
Thank u! This is a great roadmap.
Very helpful. Thanks for sharing your experiences
This was helpful. Subscribed
Thank you for the advice!
This is a very good guideline. Thank you.
Amazing sir, thanks much. Please do more.
Thanks a lot for this detailed breakdown.
finally i now undrestand machine learning well
thank you so much!!!!!!!!!!!
This is a great video. Thank you 😊
Yep😊
Thank you for your work!
Very helpful, indeed.Thank you!
I like the general approach given here, and I am familiar with many of the learning resources and recommend them. However, I do not think it is reasonable to learn all of the material listed here in seven months. That would be rushing things, to put it mildly. With each topic--math, general programming, ML stacks, ML--you would learn to complete exercises, but not how to address new sitatuions. It takes longer than seven months, in my opinion, even if you devote yourself 100%.
Hi guys! I'm starting today this roadmap. I'll be updating my journey in this comment so I don't get demotivated. See ya!
Ok, now I'm realising I just have to remember the math stuff. I'll give each course (multivariable calculus, differential equations, linear algebra and statistics probability) a week. In a month from now, I should have a strong math foundations
A week went by and I'm just starting with multivariable calculus (life has been hard). I said that I would give each math course one week, but man I love math. I will finish every course instead of just learning the basics because something tells me that I will need that information later when I finish the machine learning journey
Now that the second week has passed, I can now say that I'm concerned about the time I have spent studying. I have multivariable calculus on pause because I have to study linear algebra to understand every concept needed. Also, I didn't mention this, but I'm supposed to build an AI for a bet I did with a friend. I might jump straight to coding before finishing math to speed up things. See ya next week!
Third week completed and guess what, I'm still with maths. I've got a new hobbie (jugger) and you could say that I sure am procrastinating, statement that is true BUT progress is being made. Will I be able to learn the coding part in one month so I can study math for 2 months now? Probably not, but it's worth the risk
A month has been completed. Linear algebra is about to be finished and I'll finish multivariable calculus in a week or two. I can't wait to start coding :(
Differential equations and statistics probability haven't been touched yet, but their time will come, I guess...
Hi Patrick, I'm concerned about ( Machine Learning Courses 3:35 ) the specialization is not free unless you take courses one by one and audit them. Is that what you meant!!! Auditing the course... since auditing means I can't participate in the homework and projects??? What is your recommendation for either auditing the course and practicing from your playlist (Machine Learning from Scratch - Python Tutorials) or enrolling and paying the tuition in Coursera to fully utilize the course tools, such as projects? Thanks
you can apply financial aid, and you’ll get the course for free in 15 days
@@arunachalamm3399 use chatgpt to apply and every year I get 10 courses for free(WHICH IS MAX)
Sir, thank you for this video. Sir you are very inspiring.
Really good video for beginners in AI like me at least :)
Yes, this was helpful. Thank you very much.
Glad it was helpful!
Great advice!!
numpy is exactly what you need for mathematics. There is a book on linear algebra and Python. Not take book for Theoretical Linear Alghebra (it not for programmers)
I will love you forever because of this video.
Great Guidance, thanks a lot!!
Very clear and an informative video
This video is amazing
Great video, thank you so much.
Copy paste of another comment.
will update my progress here.
(Note to Self - How I would learn Machine Learning)
01:00 1. Math: Khan Academy
Recommended Courses:
- Multi-Variable Calculus
- Differential Equations
- Linear Algebra
- Statistics and Probability
02:00 2. Python
Recommended Courses
- FreeCodeCamp: Python in 4-Hours Full Course
- FreeCodeCamp: Intermediate Python in 6-Hours
02:37 3. Machine Learning TECH STACK
Most important Python libraries for Machine Learning, Data Science, and Data Visualization
Optional: Can be picked up later when doing the ML course.
Use for every project, which is why he recommends doing them now to build a base.
Follow a free crash course for now, pick up more advanced concepts later if needed.
- NumPy: Base for everything: Python Engineer - NumPy Crash Course Complete Tutorial
- Pandas: Data handling: Keith Gali - Complete Python Pandas Data Science Tutorial
- MatPlotLib: Visualization: FreeCodeCamp - MatPlotLib Crash Course
--------------------------- The following MachineLearning courses aren't yet needed
- Tensor Flow
- Scikit Learn
- PyCharge ???
03:35 4. Machine Learning Courses
- Machine Learning Specialization by Andrew Ng (Coursera)
- Implement algorithm from scratch using his 'ML from SCRATCH' playlist
- ML from Scratch Playlist by Python Engineer (Assembly AI)
04:45 5. Hands - On & Data Preparation
Kaggle Courses
- Intro to Machine Learning
- Intermediate Machine Learning
05:19 6. Practice & Build Portfolio
Kaggle: Competitions
- They provide lots of datasets, platform to evaluate, and a community.
06:15 7. Specialize & Create Blog
- NLP
- PyTorch / Tensor Flow
- MLOps
06:52 Start a VLOG
- Tutorial
- Share what you've learned
- Share the projects you've built
- Problems faced and how you have solved them
- Write about a topic
07:24 Books
- Machine Learning with PyTorch and SckiKit-Learn by Raschka
- Hands-On Machine Learning with SciKit-Learn, Keras & TensorFlow by Geron
I'm good at math. so, skipping that part.
I know basic python but still going through the 2nd part. starting on 29/11/23
Strong Suggestion: Make a video for "beauty in the underlying math of ML..."
Awesome incredible advice! TQ so much
Subscribed!
Thank you so much!
Math. Thank fudging god. So many of the well known tech UA-cam data scientists cover the same topic, unfortunately they are just glorified business analysts (no disrespect to BAs or DAs their profound knowledge in domain and data is what drives the industry). However, a true data scientist is a statistician/mathematician first. Programming is literally a language. I cannot be more literal when I say language, just as English , Latin, Sanskrit are languages to communicate, programming is simililar. Python has a lot of “vocabulary” (libraries),accommodating machine learning, we can implement any ml algo using any language. That’s it. If you have a proper mathematician and a proper coder that’s the ultimate combo. Design the algo code the basics, give to a proper coder and they will optimize it the compute. Cheers mate. Subbed.
Very helpful, thank you
Thanks 👍🏻
Thx, You are amazing!
This is literally what I needed to be able to start learning ML and move away from Full Stack Engineering.
Why do you want to move away from Full Stack Engineering? I'm a fresher who's being trained in full stack, and I'm wondering if it's the best way to go for me. Can you tell me what has turned you away from Full Stack?
its very helpful and thanks for the information
very helpful, thank you!
thanks for teaching us 👍
Question: For the sake of having a mentor / tutor, do you recommend taking a pricey course on, say Interview Kickstart? They have an $11,000 full AI course that has tutors who work in the industry there to help you at least 3 days a week. For the sake of organization and knowing for sure you're learning the right stuff, I can see that being a good thing, but can this stuff still be learned well enough for much less money to get a job in the industry? Would love you know your thoughts and anyone else's reading this - thanks!
In this day and age there are way to many options and opinions out there that don't make sense - it was nice to get a clear and concise short take on what to do. Even though I am a senior in college and have a decent amount of personal experience with programming and full stack web dev and some industry experience with those, I'll be following all these steps to make sure I get the best foundation possible.
Thank you for this video...
Starting on July 1st 2023 , keep liking and responding to my comment with your goals so we can all comeback here and update our progress 🙌
can you make a video elaborating how to build a more effective portfolio in ml and which platform to use?
Great information thanks man
thanks you are a king
imma start my journey here
Great video :)
Will let you know. But I’m extremely motivated 🤓
That's awesome to hear! Good luck with learning
if you don't know the background math then you are an aficionado
Thank you bruh too useful