Number 1 thing that I've learnt from Dan is that "what is the alternative" and "process will have days that will suck and that's ok" and thanks for teaching me that Dan :)
6:30 not being able to keep up is an huge demotivator. Ignoring it feels counterproductive, but it's not. 7:00 Yes! This goes with anything you're learning.
" A little everyday", an advice much needed for me. I am so stuck on my master thesis in machine learning that being so scared of not being able to solve it, I simply stopped working and the pressure simply mounted on me and I felt so bad, can't even put it into words. Now hearing your advice that I shouldn't try to control thing which are simply not in my control and rather work number of hours which are truly in my control makes way more sense. I will reach somewhere at least better than nowhere. Thanks again mate
You went above and beyond by creating your own master's degree. I'm a student of your own little self made virtual University 😂. Thank you so much for making it easier for me.
@@mrdbourke even from IITK, guys are joining online things they couldn't even think of at the time of enrolling coaching for the sake of crores of money from the placement stories. I n s p i r i n g,
Persistence is the key. It has the ability to carve rivers through rocks, But one step at a time. P.S.: I am saying this to myself more than to anybody else.
Happy Sunday y'all! One of the biggest questions I get is how to start learning machine learning. I decided to put some of my most common responses into a video. What advice would you give to people who are starting out self-learning machine learning?
Thanks,Daniel for sharing your journey and lessons. I can identify with the points you mentioned -to start building even when we are not ready, to focus on what we can control and not beating ourselves up for what we dunno. Great advices there! :)
it was a good and fresh point of view of self studying and learning, having a good balance between "putting pressure on yourself " and "taking it easy" is the key to learning well and this balance can get better when you continue struggling for it, thanks for the simple and good video, hope to watch more from you :)
Some great rules there, mate. Thanks for that. I often find myself getting frustrated when trying to rush my learning process and for not knowing things that quite simply - I haven't spent enough time on yet. As you say, a little bit every day. I find it hard to congratulate myself but all too easy to beat myself. Happy learning!
having a project and a defined end goal for that project really helped. It made doing 10-12 hours days unnoticeable. browsing stackoverflow etc for answers and solutions was a breeze. After the testing and production phases of my project I just started another.
mate, no joke, this is fucking invaluable information ! Seriously, thank you for your content, some of the most in-depth, meaningful content I have seen anywhere... holy shit, congratulations lad, stellar stuff.
"you know how much you don't know so you think you suck even more". So true, story of my life. Even knowing that, it's hard for me getting past my imposter syndrome.
The analogy between a finished ML project and a finished movie was great. There are countless of work behind any successful model or project. Thank you also for the resources you recommended. I hadn't known about any of them. I'll check them out.
Yeah you are really right, there's lots of new things going around, that I want to learn or do, but it'll be of no use, and at last I have to let go some of them and concerntrate on my learning and academics.
The best advice I got so far on ML. I've been studying python for 2 months and because of math and my curiosity I got offered a machine learning research engineer position, Now I'm working as a full time research engineer and study at night till 12am and on weekend. but still needed some guidance and this was it! Thanks bro!
Hi Daniel, I really have to thank you for this video. Surprisingly I've just received it as a recommendation on Medium's Digest despite being 2 years old. I've finished my master's in AI in 2016 and lost my way since then. Last year I started my doctorate but without too much expectation. Every now and then I get negative especially when I see so many young data scientists tackling machine learning problems, mathematical concepts and statistical questions so easily. Your speech was precise on what we need to listen to and do not forget on this learning journey. Thank you again, I really appreciate it!
Awesome man! Great to hear your path has worked out for you, and also fantastic to see the large community of people interested in self learning - it really is motivating!
One thing I actually wanted to ask you - I've been self teaching maths for the past year or so (didn't do well in highschool) so started with algebra then moved to precalculus and now tackling calculus and linear algebra bit by bit. How have you dealt with brushing up on your mathematics? And do you feel it to be an important skill to have working as an ML engineer?
I used Khan Academy to learn math almost exclusively. There are a few other resources such as the math for machine learning book. My philosophy is to learn enough to get through the next roadblock of whatever project I’m working on. Trying to learn it all at one go is like trying to boil the ocean. My role was 99% code-based. The math was abstracted away with libraries such as TensorFlow and PyTorch.
These are some very valuable points, also, i would add one which helped me a lot is that keep taking breaks, take breaks everyday, every week, every month, and kind of view the process you are making, be grateful for that you have learnt. Its so easy to loose your mental health, because the journey is very long, and you see people around you building some awesome projects. So, do things everyday which make you feel alive, for me its sitting in the nature and reading books, and as i am indian, i am quite spiritual, so, i medidate, and chant mantras too. These are things which really make me feel alive.
Hello there! We are researchers in human-computer interaction (HCI) looking for people who have taken an initiative to recently learn Machine Learning on their own, for career, course or curiosity. Would you mind telling us here (www.surveymonkey.ca/r/SelfLearning_ML) about your experiences and any difficulties you faced while self-teaching ML and how you overcame them. There is also a chance to win $50 giftcard. You can help this project by taking out 5-10 minutes to participate in our study. For more details, see here: www.surveymonkey.ca/r/SelfLearning_ML Please share this request with your colleagues or friends who fit this description. People from any major/background may participate. The survey will be open until July 31, 2020.
This was amazing advice.This is the first time someone convinced me to slow down my aspirations and speed up my actual learning. I did some deep RL projects and started to want to go to OpenAI and DeepMind to work on AGI and AI Safety which I am passionate about. But I realize how far I am in my knowledge in order to be able to contribute or even understand anything properly. I'll take it patiently from now on, and try to actually reach there rather than rewarding my brain falsely with hollow dreams that I have no idea how I'll fulfill. I don't even have a CS degree so I won't even be able to do an AI Masters. I'll just try to learn from my own country.
This is good! I am trying to do Machine Learning! I am trying to motivate myself to learn something on my own! I struggling with consistency, discipline, diligence and the only thing that I constantly doing is procrastinating. Thanks to this video. This is truly a motivation!
This is exactly what I needed to hear right now. Feel like I could end up going on a never ending mooc spiral, so point 3 was what I needed. Thanks dude!
That’s what I did! The MOOC spiral! Great way to put it. I should’ve started on my own things sooner but I was scared... scared of failing. Which is often how you learn best. Keep learning Amarjeet!
You're a legend mate. Just started Python about two weeks ago, almost finished the basics on w3schools. I needed money and was going to try make a youtube channel, but my voice is not great so i started thinking of a ways to impove that. Voice cloning my favorite voice actor was the idea I came up with. The only free resource to do this required Python. This introduced me to Python and my interest in machine learning was sparked. I would much rather be a coder than a professional speaker, so I started studying coding.
about 85% through the DataCamp intermediate programme (data analyst) and loving it. absolutely blows treehouse, Dataquest and youtube tutorials out of the water as far as I'm concerned. obviously will be different for everyone though
@@mrdbourke definitely worth having a look they have a range of subjects all the way through deep learning and statistical thinking... all the way through to applied finance and reporting. and it isn't just in python, you can also do R, Git & SQL.
Sometimes for me, watching videos is simply reminding myself that there are a lot of people still working on machine learning out there and i need to do more and more.
Thanks for making your videos. You never know when someone is going find them and run with them. I’m just restarting my ML journey after many false starts. This time, with the resources that you have shared, I feel that I can take it to a successful conclusion. 👍🙏
Thanks for this video Daniel, today's the day I am feeling frustrated with myself and now I know what I got to do - embrace the suck and keep moving forward
Definitely the best piece of advice I'll get all week long! You've made me realise how I've been trying to rush my ML learning journey ( just as I normally do when I'm learning a new web framework actually!) which is probably not going to be beneficial as I thought it would! I've been trying to learn a ton of stuff all at once, which I'm realising is not the best way to master ML.Thanks!
Hi Elvis! We are researchers in human-computer interaction (HCI) looking for people who have taken an initiative to recently learn Machine Learning on their own, for career, course or curiosity. Would you mind telling us here (www.surveymonkey.ca/r/SelfLearning_ML) about your experiences and any difficulties you faced while self-teaching ML and how you overcame them. There is also a chance to win $50 giftcard. You can help this project by taking out 5-10 minutes to participate in our study. For more details, see here: www.surveymonkey.ca/r/SelfLearning_ML Please share this request with your colleagues or friends who fit this description. People from any major/background may participate. The survey will be open until July 31, 2020.
Point number 4 struck me the hardest Yes, new papers are cool and all and will give insights, but in the end, the knowledge will go away anyways if i dont apply them Thanks for this daniel, time to shove all those tabs to the bookmark
It’s rly not an easy journey but the most I need is rly more motivation and focus. Attention is rly rly the key to anything said. Must concentrate and learn the concepts one by one. How to stay focused?
4,5,7point related to me, today I frustrated because someone advised me that I can't do it, or should stop this course(family:) , I have a issues of money like everyone :), I like ml/ai more than any other things in the world but today people telling me that I can't, my father don't have money problem but I have;) I like self studies and I trying to learn ml or ai too fast,I did python in 1 month I am at intermediate level , ur video helped me thanks 👍 Edit = [numpy, pandas, ml(classification, regression algorithms ), scikit learn overview] that's all I did, if u have any suggestion than please 👇 .
Goog tips!! But my goal is to complete 2 adequate projects by May and end up that micro-stage as a ML intern in one local company. Hope to report here later about the results!! Time limits may help anyway
I mean. Thank you so much I can't thank you enough. I've been getting really anxious lately looking at how much I don't know and all my friends are getting jobs and promotion while I'm still studying. I know what I should be doing, I just kind of loose sight of the bigger picture sometimes.
Hi Eza, We are researchers in human-computer interaction (HCI) looking for people who have taken an initiative to recently learn Machine Learning on their own, for career, course or curiosity. Would you mind telling us here (www.surveymonkey.ca/r/SelfLearning_ML) about your experiences and any difficulties you faced while self-teaching ML and how you overcame them. There is also a chance to win $50 giftcard. You can help this project by taking out 5-10 minutes to participate in our study. For more details, see here: www.surveymonkey.ca/r/SelfLearning_ML Please share this request with your colleagues or friends who fit this description. People from any major/background may participate. The survey will be open until July 31, 2020.
In addition to make progress everyday, make studying a habit, an essential part of your day. Aim to study at the same time each day and at the same cadence each week.
if i remember correctly Newton was paraphrasing a 14th century astronomer with that quote. Also there is a belief that hewasnt being self depricating but rather was making a sly remark against Hooke because he was short and they very very much hated each other.
Yo Srinivas, I’m not the best person to answer this, I haven’t applied for a job here in 2-3 years I’d spend some time doing research on different companies which would benefit from your skills
Daniel - Great video! I come from a bit more formal ML background, and I am definitely looking to find more resources on NLP. Any suggestions? Also, I am trying to expand my data science RUclips presence. Do you have any tips for engaging with the community?
Yo Ken! Thank you for the kind words. For state of the art NLP, I've been loving Flair: github.com/zalandoresearch/flair But for learning NLP, I'd suggest Standford's NLP course on RUclips: ruclips.net/video/OQQ-W_63UgQ/видео.html Or There's a Coursera specialization called Natural Language Processing which I've heard good things about (but haven't done), it has been on my radar: www.coursera.org/learn/language-processing
And for RUclips I don’t really have secrets or hacks other than to make videos you’d like to watch, chances are someone else out there probably does too
Thanks Daniel I've learnt python and did basics of numpy,pandas,matplotlib and seaborn but I haven't started ml so far so should I look up for some projects at this moment or learn ml first?
Hello sir I’m actually a masters degree student in applied statistics in China and my field of research is predictions using machine learning concepts. So basically I don’t know where to start and how to make really efficient my learning ? I’ve basic knowledge in python and r as well. My dream is to get a job as data scientist.
That’s awesome Franck! One way that helps me make my learning clear for each new day is to write down a list of things I want to accomplish every morning and working through them. For example, “Today I’m going to study for 3 hours” then I’ll have check boxes for every hour
Hey, is it possible for a 16-year-old (if the 16-year-old has done studies in P.H..D level by himself without going to a college )to get a job in AI? If possible how to gate a chance for an interview
If you have that level of skills and the ability to communicate them, finding a job shouldn't be too hard. Expect a response rate of 1 in 100 if you apply online. Best to reach out to people and start making relationships in the field. Your age shouldn't matter but some people may be turned off by it (it's nothing against you, only reality). If there are events on the field in your area, you should attend them.
What is your opinion on learning linear algebra , probability and calculus? top down or bottom up? Ive only started Andrew Ng's coursera course, but foreseeably this knowledge would be necessary to truly understand ,choose and tweak the models ?
Hello there! We are researchers in human-computer interaction (HCI) looking for people who have taken an initiative to recently learn Machine Learning on their own, for career, course or curiosity. Would you mind telling us here (www.surveymonkey.ca/r/SelfLearning_ML) about your experiences and any difficulties you faced while self-teaching ML and how you overcame them. There is also a chance to win $50 giftcard. You can help this project by taking out 5-10 minutes to participate in our study. For more details, see here: www.surveymonkey.ca/r/SelfLearning_ML Please share this request with your colleagues or friends who fit this description. People from any major/background may participate. The survey will be open until July 31, 2020.
Hi Daniel, Thanks for the information. I have a MSc in Mathematics and Computing. The easiest part for me in ML is mathematics. Will this be beneficial for me to land in a ML job? My programming skills are not as great as any CS guy but I am currently working on it. Thanks again 😀
If you know the ML math, use that to your advantage, start creating projects that showcase your knowledge - then they can be used as showing someone "here's what I've done" when you go for a job
Whichever way suits you best. With enough practice, you will find it. Otherwise, Coursera’s learning how to learn course is a great way to find different learning principles which can help
Is there such thing as 'junior machine learning engineer'? How long can it take for someone, who is just starting learning to code and coming from totally different background, to be able to get a ML job, if they study for 2 hours per day? Is it a good idea to start first with back end Python developer job, or one can get straight into ML, without much longer study period?
Hey Nikolay, there is such a thing. I was once a junior machine learning engineer (I’d still count myself as one). I started learning code (and ML code) and got a job 9-months later. But it really depends. I was studying 4-6 hours per day. So 2-hours may take longer to get the baseline skills needed for a role. This article may help: www.mrdbourke.com/startaiml/
@@mrdbourke Ok, I read it. It's helpful, thanks. I'm doing Andrei's Python Zero to Mastery course on udemy and that's how I've heard about you - from their Discord group. I am interested in AI research and the potential of creating AGI, but will be good probably to get some Python developer job at first, as soon as possible. And then I can continue with the AI learning.
Hi Nikolay, We are researchers in human-computer interaction (HCI) looking for people who have taken an initiative to recently learn Machine Learning on their own, for career, course or curiosity. Would you mind telling us here (www.surveymonkey.ca/r/SelfLearning_ML) about your experiences and any difficulties you faced while self-teaching ML and how you overcame them. There is also a chance to win $50 giftcard. You can help this project by taking out 5-10 minutes to participate in our study. For more details, see here: www.surveymonkey.ca/r/SelfLearning_ML Please share this request with your colleagues or friends who fit this description. People from any major/background may participate. The survey will be open until July 31, 2020.
Hi @@RoyalBengalCub :) . Thank you for the attention. I will check out your survey. But I'm still learning Python, I haven't reached to ML yet, so I'm probably not the right fit for your research. It sounds very interesting though...
Interesting that you say to start building things when you're 70% ready. I've read some writing from Jezz Bezos (Amazon founder), he says to make a decision when you are 70% sure of something. If you wait until longer than that, then you've waited too long. I thought this was an interesting parallel.
@@mrdbourke Yes, it was actually in one of his annual shareholder letters to investors. Thanks for the vid, I've watched a few of yours in the last few days.
@@BluEN1111they contributed to me starting a career in ML but no resource fully prepares you. Depending on where you work, you'll probably find most of the learning happens on the job. Learning never stops. Get a foundation of knowledge and then get used to using it on your own projects, this will be helpful when starting your career.
Another video, sweet. On a side note, I uploaded a video on a new channel I started, was wondering if you can give me some tips on improving. Edit: Have you checked out the new Andrew NG AI for everyone course?
You're quick Paul! Sure thing, send it through to me on telegram: t.me/mrdbourke or email. The Andrew Ng course -- I have seen it! My brother is probably going to do it, he's in the business world but not a developer yet.
Hey Daniel, I am new in technical line. I want to become an Ai engineer. Can you please tell me what to do should I study first Machine learning or data science or Ai
Start with whichever interests you most, you’ll start to figure out which one is which, they have many similarities Data science and ML can be considered subsets of AI
These advice are impractical because your boss expect you to master ML in a week, or even in two days. Such bosses can be found easily in almost every university, mainly Chinese assistant professors, most of whom think you were born an ML expert.
You can earn money through a machine learning job (freelance or for a company) or by using your skills to build something people can use and charging them for it.
Number 1 thing that I've learnt from Dan is that "what is the alternative" and "process will have days that will suck and that's ok" and thanks for teaching me that Dan :)
Embrace the suck.
6:30 not being able to keep up is an huge demotivator. Ignoring it feels counterproductive, but it's not. 7:00 Yes! This goes with anything you're learning.
" A little everyday", an advice much needed for me. I am so stuck on my master thesis in machine learning that being so scared of not being able to solve it, I simply stopped working and the pressure simply mounted on me and I felt so bad, can't even put it into words. Now hearing your advice that I shouldn't try to control thing which are simply not in my control and rather work number of hours which are truly in my control makes way more sense. I will reach somewhere at least better than nowhere. Thanks again mate
You went above and beyond by creating your own master's degree. I'm a student of your own little self made virtual University 😂. Thank you so much for making it easier for me.
If the road ahead doesn’t suit you, make you’re own
@@mrdbourke even from IITK, guys are joining online things they couldn't even think of at the time of enrolling coaching for the sake of crores of money from the placement stories. I n s p i r i n g,
Persistence is the key. It has the ability to carve rivers through rocks, But one step at a time.
P.S.: I am saying this to myself more than to anybody else.
Most of this video was notes to myself as well :)
Happy Sunday y'all! One of the biggest questions I get is how to start learning machine learning. I decided to put some of my most common responses into a video.
What advice would you give to people who are starting out self-learning machine learning?
Thanks,Daniel for sharing your journey and lessons. I can identify with the points you mentioned -to start building even when we are not ready, to focus on what we can control and not beating ourselves up for what we dunno. Great advices there! :)
it was a good and fresh point of view of self studying and learning, having a good balance between "putting pressure on yourself " and "taking it easy" is the key to learning well and this balance can get better when you continue struggling for it,
thanks for the simple and good video, hope to watch more from you
:)
Thank you Daniel, your AI masters degree helped a lot. Your motivation to stay fit and keep learning is one of the underrated stuffs. Thank you again.
Thank you Debadatta! That’s really kind of you to say! All the best. #keepmoving #keeplearning
Some great rules there, mate. Thanks for that. I often find myself getting frustrated when trying to rush my learning process and for not knowing things that quite simply - I haven't spent enough time on yet. As you say, a little bit every day. I find it hard to congratulate myself but all too easy to beat myself. Happy learning!
Too right. I do the same. Keep learning concrete! Thanks for the kind words
having a project and a defined end goal for that project really helped. It made doing 10-12 hours days unnoticeable. browsing stackoverflow etc for answers and solutions was a breeze. After the testing and production phases of my project I just started another.
mate, no joke, this is fucking invaluable information ! Seriously, thank you for your content, some of the most in-depth, meaningful content I have seen anywhere... holy shit, congratulations lad, stellar stuff.
Thank you thank you thank you, so stoked you enjoyed
"you know how much you don't know so you think you suck even more". So true, story of my life. Even knowing that, it's hard for me getting past my imposter syndrome.
The analogy between a finished ML project and a finished movie was great. There are countless of work behind any successful model or project. Thank you also for the resources you recommended. I hadn't known about any of them. I'll check them out.
Thanks Bianca! I'm glad you liked it :) All the best with the resources!
Yeah you are really right, there's lots of new things going around, that I want to learn or do, but it'll be of no use, and at last I have to let go some of them and concerntrate on my learning and academics.
Sometimes new work research doesn’t get taken advantage of for years, it’s only until it can be applied. But we have a bias to be focused on the new
The best advice I got so far on ML. I've been studying python for 2 months and because of math and my curiosity I got offered a machine learning research engineer position, Now I'm working as a full time research engineer and study at night till 12am and on weekend. but still needed some guidance and this was it! Thanks bro!
thats cool, do you have a degree?
Hi Daniel, I really have to thank you for this video. Surprisingly I've just received it as a recommendation on Medium's Digest despite being 2 years old. I've finished my master's in AI in 2016 and lost my way since then. Last year I started my doctorate but without too much expectation. Every now and then I get negative especially when I see so many young data scientists tackling machine learning problems, mathematical concepts and statistical questions so easily. Your speech was precise on what we need to listen to and do not forget on this learning journey. Thank you again, I really appreciate it!
Awesome man! Great to hear your path has worked out for you, and also fantastic to see the large community of people interested in self learning - it really is motivating!
One thing I actually wanted to ask you - I've been self teaching maths for the past year or so (didn't do well in highschool) so started with algebra then moved to precalculus and now tackling calculus and linear algebra bit by bit. How have you dealt with brushing up on your mathematics? And do you feel it to be an important skill to have working as an ML engineer?
I used Khan Academy to learn math almost exclusively. There are a few other resources such as the math for machine learning book. My philosophy is to learn enough to get through the next roadblock of whatever project I’m working on. Trying to learn it all at one go is like trying to boil the ocean. My role was 99% code-based. The math was abstracted away with libraries such as TensorFlow and PyTorch.
These are some very valuable points, also, i would add one which helped me a lot is that keep taking breaks, take breaks everyday, every week, every month, and kind of view the process you are making, be grateful for that you have learnt. Its so easy to loose your mental health, because the journey is very long, and you see people around you building some awesome projects.
So, do things everyday which make you feel alive, for me its sitting in the nature and reading books, and as i am indian, i am quite spiritual, so, i medidate, and chant mantras too. These are things which really make me feel alive.
Hello there!
We are researchers in human-computer interaction (HCI) looking for people who have taken an
initiative to recently learn Machine Learning on their own, for career, course or curiosity. Would you mind telling us here (www.surveymonkey.ca/r/SelfLearning_ML) about your experiences and any difficulties you faced while self-teaching ML and how you overcame them. There is also a chance to win $50 giftcard.
You can help this project by taking out 5-10 minutes to participate in our study.
For more details, see here: www.surveymonkey.ca/r/SelfLearning_ML
Please share this request with your colleagues or friends who fit this description. People from any major/background may participate. The survey will be open until July 31, 2020.
Thanks for sharing, myself have gone through all these questions, comforting to know overall struggle is the same :),
Thanks!
This was amazing advice.This is the first time someone convinced me to slow down my aspirations and speed up my actual learning. I did some deep RL projects and started to want to go to OpenAI and DeepMind to work on AGI and AI Safety which I am passionate about. But I realize how far I am in my knowledge in order to be able to contribute or even understand anything properly. I'll take it patiently from now on, and try to actually reach there rather than rewarding my brain falsely with hollow dreams that I have no idea how I'll fulfill. I don't even have a CS degree so I won't even be able to do an AI Masters. I'll just try to learn from my own country.
Great insights Akarsh, keep enjoying it my friend.
This is good! I am trying to do Machine Learning! I am trying to motivate myself to learn something on my own! I struggling with consistency, discipline, diligence and the only thing that I constantly doing is procrastinating. Thanks to this video. This is truly a motivation!
This is exactly what I needed to hear right now. Feel like I could end up going on a never ending mooc spiral, so point 3 was what I needed. Thanks dude!
That’s what I did! The MOOC spiral! Great way to put it. I should’ve started on my own things sooner but I was scared... scared of failing. Which is often how you learn best. Keep learning Amarjeet!
You're a legend mate. Just started Python about two weeks ago, almost finished the basics on w3schools. I needed money and was going to try make a youtube channel, but my voice is not great so i started thinking of a ways to impove that. Voice cloning my favorite voice actor was the idea I came up with. The only free resource to do this required Python. This introduced me to Python and my interest in machine learning was sparked. I would much rather be a coder than a professional speaker, so I started studying coding.
You got this brother! Love the idea!
about 85% through the DataCamp intermediate programme (data analyst) and loving it. absolutely blows treehouse, Dataquest and youtube tutorials out of the water as far as I'm concerned. obviously will be different for everyone though
That’s great to hear! I haven’t used it too much but my brother is loving it. I’ll note this for next time someone asks where to start
@@mrdbourke definitely worth having a look they have a range of subjects all the way through deep learning and statistical thinking... all the way through to applied finance and reporting. and it isn't just in python, you can also do R, Git & SQL.
This is a great and insightful video, one of the greatest I've seen on the subject. Something you can rewatch whenever self-doubt kicks in.
Thanks Sasha!
really simple stuff but usually overlooked, thanks for sharing
Where have this video been all this time
Watched it in 2020 and totally wonderful advices , keep it up sir 👍
The tip to don't try to control things you can't control was very useful to me thx
Sometimes for me, watching videos is simply reminding myself that there are a lot of people still working on machine learning out there and i need to do more and more.
Thanks for making your videos. You never know when someone is going find them and run with them. I’m just restarting my ML journey after many false starts. This time, with the resources that you have shared, I feel that I can take it to a successful conclusion. 👍🙏
Thanks man, needed to hear this to bring back some perspective.
Thanks for this video Daniel, today's the day I am feeling frustrated with myself and now I know what I got to do - embrace the suck and keep moving forward
Keep getting after it Benjamin!
Thank you! Very helpful advices 🎉
Definitely the best piece of advice I'll get all week long! You've made me realise how I've been trying to rush my ML learning journey ( just as I normally do when I'm learning a new web framework actually!) which is probably not going to be beneficial as I thought it would! I've been trying to learn a ton of stuff all at once, which I'm realising is not the best way to master ML.Thanks!
Hi Elvis!
We are researchers in human-computer interaction (HCI) looking for people who have taken an
initiative to recently learn Machine Learning on their own, for career, course or curiosity. Would you mind telling us here (www.surveymonkey.ca/r/SelfLearning_ML) about your experiences and any difficulties you faced while self-teaching ML and how you overcame them. There is also a chance to win $50 giftcard.
You can help this project by taking out 5-10 minutes to participate in our study.
For more details, see here: www.surveymonkey.ca/r/SelfLearning_ML
Please share this request with your colleagues or friends who fit this description. People from any major/background may participate. The survey will be open until July 31, 2020.
Point number 4 struck me the hardest
Yes, new papers are cool and all and will give insights, but in the end, the knowledge will go away anyways if i dont apply them
Thanks for this daniel, time to shove all those tabs to the bookmark
Keep at it Hasan! 👊
Thats exactly what i needed to hear! Thanks from Brazil!! 🙏
Very helpful! N. 6 is my big problem. I need to take it easy on me! Nice channel!
We’re all our own worst enemies sometimes!
Thanks Diário :)
It’s rly not an easy journey but the most I need is rly more motivation and focus. Attention is rly rly the key to anything said. Must concentrate and learn the concepts one by one. How to stay focused?
4,5,7point related to me, today I frustrated because someone advised me that I can't do it, or should stop this course(family:) , I have a issues of money like everyone :), I like ml/ai more than any other things in the world but today people telling me that I can't, my father don't have money problem but I have;) I like self studies and I trying to learn ml or ai too fast,I did python in 1 month I am at intermediate level ,
ur video helped me thanks 👍
Edit = [numpy, pandas, ml(classification, regression algorithms ), scikit learn overview] that's all I did, if u have any suggestion than please 👇 .
Awesome piece of advise Daniel
Thanks a lot man, gave me a lot of confidence
No worries Jayanth! Glad you enjoyed it!
You can also think about making fitness videos as well. Great tips for ML by the way.
Hey Daniel, you should make Self-Studying Machine Learning? resources for 2020. I am telling you it will be a hit.
Great idea my friend, stay tuned
Goog tips!! But my goal is to complete 2 adequate projects by May and end up that micro-stage as a ML intern in one local company. Hope to report here later about the results!! Time limits may help anyway
Massive effort! Let me know how you go :)
I mean. Thank you so much I can't thank you enough. I've been getting really anxious lately looking at how much I don't know and all my friends are getting jobs and promotion while I'm still studying. I know what I should be doing, I just kind of loose sight of the bigger picture sometimes.
You got this Abhishek! Keep learning my friend
Thanks a lot , man. So honest. Very helpful.
Of course Eza, glad you enjoyed
Hi Eza,
We are researchers in human-computer interaction (HCI) looking for people who have taken an
initiative to recently learn Machine Learning on their own, for career, course or curiosity. Would you mind telling us here (www.surveymonkey.ca/r/SelfLearning_ML) about your experiences and any difficulties you faced while self-teaching ML and how you overcame them. There is also a chance to win $50 giftcard.
You can help this project by taking out 5-10 minutes to participate in our study.
For more details, see here: www.surveymonkey.ca/r/SelfLearning_ML
Please share this request with your colleagues or friends who fit this description. People from any major/background may participate. The survey will be open until July 31, 2020.
Thank you I needed this more than you realize right now.
Glad you found value Jacob! Keep getting after it brother.
Awesome video. Going to read your articles on medium 😃.
Thanks Yash!
In addition to make progress everyday, make studying a habit, an essential part of your day. Aim to study at the same time each day and at the same cadence each week.
Great point Carlo. Habits are powerful. Pretty sure they’re in a different part of the brain to other things. Or maybe that’s something else...
Ressszzz
if i remember correctly Newton was paraphrasing a 14th century astronomer with that quote. Also there is a belief that hewasnt being self depricating but rather was making a sly remark against Hooke because he was short and they very very much hated each other.
Yes, I have also heard that it was an insult at Robert Hooke.
The second graph does not pass the vertical line test... What a pity... 😅
In the background what is black table kind of thing that your computer is kept on ? @DanielBourke
It’s called a varidesk! It’s a desk that changes height so you can stand and sit
Nice Video Dude , Related to it completely .
Thank you Nayan, great to hear!
@@mrdbourke Hey man , are you into Web Development ? i am Kinda stuck , it would be really helpful if you could give me some advice.
@@nayansinghai6593 sure thing, contact me any time: daniel (at) mrdbourke (dot) com
Thanks, dude, you are my motivation :)
Keep getting after it Lucky 👊
Ohh here he is. The Saviour💛
Excellent advice, thanks for sharing
Glad you enjoyed Darren!
Thankyou bro.. great motivation.....:)
04:00 I kid u not, what u just said there triggers my phone Google to search for Learning ML...
very impressive and helpful!
Thanks Chan! Glad you enjoyed :)
Great post !!
Thanks Jebastin!
Hey Daniel, how is the job market for beginners in DS and ML in Australia?
Yo Srinivas, I’m not the best person to answer this, I haven’t applied for a job here in 2-3 years
I’d spend some time doing research on different companies which would benefit from your skills
I find this so encouraging. Truly, we need to understand first the math of machine learning before going into the coding.
I disagree. It has to be both at the same time. Id recommend starting with coding if anything
Daniel - Great video! I come from a bit more formal ML background, and I am definitely looking to find more resources on NLP. Any suggestions?
Also, I am trying to expand my data science RUclips presence. Do you have any tips for engaging with the community?
Yo Ken! Thank you for the kind words. For state of the art NLP, I've been loving Flair: github.com/zalandoresearch/flair
But for learning NLP, I'd suggest Standford's NLP course on RUclips: ruclips.net/video/OQQ-W_63UgQ/видео.html
Or
There's a Coursera specialization called Natural Language Processing which I've heard good things about (but haven't done), it has been on my radar: www.coursera.org/learn/language-processing
And for RUclips I don’t really have secrets or hacks other than to make videos you’d like to watch, chances are someone else out there probably does too
@@mrdbourke Thank you! I will check it out!
@@mrdbourke Great! Doing my best on that angle. I am having a lot of fun making videos, still trying to flesh out my on-camera personality
How? By stopping watching this, and starting to watch some nice lectures. ;)
How? By stopping to watch lectures and starting to get your hands dirty building projects
@Saqif Mohammed how? By not replying to comments and waste your time
@@Osama-kx3cq 😂
Thanks Daniel I've learnt python and did basics of numpy,pandas,matplotlib and seaborn but I haven't started ml so far so should I look up for some projects at this moment or learn ml first?
Hey Dan pls make a video on how to get a job as machine learning engineer for self-thaught
Great idea! I wrote an article on this which may help in the meantime: bit.ly/3thingsforML
I want to learn how apply machine learning to Seismic Data
Great video!
THANK YOU✌️
It is indeed worth it
Daniel..if a feature is classified as ordinal but with numeric representation, should we apply one hot encoder to it either ?
Depends on the ordinality. If it’s truly ordinal, keep it that way, if it could be split into categories, maybe one hot is better
Patience is key
You are living my dream life
Keep getting after it Kevin 👊
Hello sir I’m actually a masters degree student in applied statistics in China and my field of research is predictions using machine learning concepts. So basically I don’t know where to start and how to make really efficient my learning ? I’ve basic knowledge in python and r as well. My dream is to get a job as data scientist.
That’s awesome Franck! One way that helps me make my learning clear for each new day is to write down a list of things I want to accomplish every morning and working through them. For example, “Today I’m going to study for 3 hours” then I’ll have check boxes for every hour
I really understand and I already fixed a 4 hours study plan everyday .
Hey, is it possible for a 16-year-old (if the 16-year-old has done studies in P.H..D level by himself without going to a college )to get a job in AI? If possible how to gate a chance for an interview
If you have that level of skills and the ability to communicate them, finding a job shouldn't be too hard. Expect a response rate of 1 in 100 if you apply online. Best to reach out to people and start making relationships in the field. Your age shouldn't matter but some people may be turned off by it (it's nothing against you, only reality). If there are events on the field in your area, you should attend them.
Thanlyou
What is your opinion on learning linear algebra , probability and calculus? top down or bottom up? Ive only started Andrew Ng's coursera course, but foreseeably this knowledge would be necessary to truly understand ,choose and tweak the models ?
Learn what you need to learn when you need to learn it. Trying to learn all of those is like trying to boil the ocean
Thank you for sharing this! I love the content, What you think about machine learning with JavaScript?
do i need very strong data structures & algorithm?
Hi Daniel, great tips, thanks for the video. Also what video editing software do you use for your channel
Glad you enjoyed Nkosinathi!
I use Final Cut Pro for all my editing :)
Sir Please suggest me resources for Time Series Forecasting in Python Please Sir
Great advices Dan!
I’m in the beginning 🛤
Thanks for the motivation ❤️🙏
Hello there!
We are researchers in human-computer interaction (HCI) looking for people who have taken an
initiative to recently learn Machine Learning on their own, for career, course or curiosity. Would you mind telling us here (www.surveymonkey.ca/r/SelfLearning_ML) about your experiences and any difficulties you faced while self-teaching ML and how you overcame them. There is also a chance to win $50 giftcard.
You can help this project by taking out 5-10 minutes to participate in our study.
For more details, see here: www.surveymonkey.ca/r/SelfLearning_ML
Please share this request with your colleagues or friends who fit this description. People from any major/background may participate. The survey will be open until July 31, 2020.
Your channel is soo underrated 💔
Thank you thank you!
Thank You
You’re welcome my friend
Hi Daniel, Thanks for the information. I have a MSc in Mathematics and Computing. The easiest part for me in ML is mathematics. Will this be beneficial for me to land in a ML job? My programming skills are not as great as any CS guy but I am currently working on it. Thanks again 😀
If you know the ML math, use that to your advantage, start creating projects that showcase your knowledge - then they can be used as showing someone "here's what I've done" when you go for a job
Great info.
Thank you Ken!
love love love Like seriously love.Thank you from the core like seriously .May God Bless you .You are awesome
Thanks Dan what's the best way to learn maths for Ai and Ml
Whichever way suits you best. With enough practice, you will find it. Otherwise, Coursera’s learning how to learn course is a great way to find different learning principles which can help
Is there such thing as 'junior machine learning engineer'? How long can it take for someone, who is just starting learning to code and coming from totally different background, to be able to get a ML job, if they study for 2 hours per day? Is it a good idea to start first with back end Python developer job, or one can get straight into ML, without much longer study period?
Hey Nikolay, there is such a thing. I was once a junior machine learning engineer (I’d still count myself as one). I started learning code (and ML code) and got a job 9-months later. But it really depends. I was studying 4-6 hours per day. So 2-hours may take longer to get the baseline skills needed for a role. This article may help: www.mrdbourke.com/startaiml/
@@mrdbourke Thanks very much, I'll check it out!
@@mrdbourke Ok, I read it. It's helpful, thanks. I'm doing Andrei's Python Zero to Mastery course on udemy and that's how I've heard about you - from their Discord group. I am interested in AI research and the potential of creating AGI, but will be good probably to get some Python developer job at first, as soon as possible. And then I can continue with the AI learning.
Hi Nikolay,
We are researchers in human-computer interaction (HCI) looking for people who have taken an
initiative to recently learn Machine Learning on their own, for career, course or curiosity. Would you mind telling us here (www.surveymonkey.ca/r/SelfLearning_ML) about your experiences and any difficulties you faced while self-teaching ML and how you overcame them. There is also a chance to win $50 giftcard.
You can help this project by taking out 5-10 minutes to participate in our study.
For more details, see here: www.surveymonkey.ca/r/SelfLearning_ML
Please share this request with your colleagues or friends who fit this description. People from any major/background may participate. The survey will be open until July 31, 2020.
Hi @@RoyalBengalCub :) . Thank you for the attention. I will check out your survey. But I'm still learning Python, I haven't reached to ML yet, so I'm probably not the right fit for your research. It sounds very interesting though...
Interesting that you say to start building things when you're 70% ready.
I've read some writing from Jezz Bezos (Amazon founder), he says to make a decision when you are 70% sure of something. If you wait until longer than that, then you've waited too long.
I thought this was an interesting parallel.
I remember reading the same thing. It’s such a good saying I had to steal it.
@@mrdbourke Yes, it was actually in one of his annual shareholder letters to investors. Thanks for the vid, I've watched a few of yours in the last few days.
Daniel, how important are algorithms and data structures for a ML job?
It’s hard to say. It depends on what you’ll be working on.
They haven’t come into play very much in my role but I’m working on getting better at them
@@mrdbourke Do you think those 3 resources could have been enough for you to start a career in ML?
@@BluEN1111they contributed to me starting a career in ML but no resource fully prepares you. Depending on where you work, you'll probably find most of the learning happens on the job. Learning never stops. Get a foundation of knowledge and then get used to using it on your own projects, this will be helpful when starting your career.
Daniel Bourke Thanks alot!
Another video, sweet. On a side note, I uploaded a video on a new channel I started, was wondering if you can give me some tips on improving.
Edit: Have you checked out the new Andrew NG AI for everyone course?
You're quick Paul! Sure thing, send it through to me on telegram: t.me/mrdbourke or email.
The Andrew Ng course -- I have seen it! My brother is probably going to do it, he's in the business world but not a developer yet.
@@mrdbourke Sweet thanks!
Hey Daniel, I am new in technical line. I want to become an Ai engineer. Can you please tell me what to do should I study first Machine learning or data science or Ai
Start with whichever interests you most, you’ll start to figure out which one is which, they have many similarities
Data science and ML can be considered subsets of AI
These advice are impractical because your boss expect you to master ML in a week, or even in two days. Such bosses can be found easily in almost every university, mainly Chinese assistant professors, most of whom think you were born an ML expert.
Can I learn python and machine learning for my thesis within 4 and a half months? Is it easy to do this while working?
You can learn it but it depends what you need to use it for, I’d advise not to rush, worthwhile skills take time to build
@@mrdbourke so like how many months will you suggest please? My background is in electrical engineering
I have a question sir.will you answer?
is SAS machine learning course good?
Jisos!!
you sound like cool teachers at school.
How to earn money after completing machine learning course self
You can earn money through a machine learning job (freelance or for a company) or by using your skills to build something people can use and charging them for it.
@@mrdbourke how to apply for job as a fresher i think it is very difficult for fresher to get a job
came here throught quora can i get your linked in profile to get in touch
Yo Sai! My LinkedIn is: linkedin.com/in/mrdbourke
Or if you search Daniel Bourke LinkedIn you should find me!
after that you can apply for internship ,,?
You can apply for an internship at any stage. But building skills is important. My advice: keep applying and keep learning at the same time.
@@mrdbourke thanks you inspire me