How Iā€™d learn ML in 2024 (if I could start over)

By
09rupeshraj4@gmail.com
-
47
39


Become better at machine learning in 5 min/ week šŸ‘‰šŸ»

In this video, I share how I would learn Machine Learning in 2024 if I could start over.
For the past 3 years, I have been studying machine learning (and 2 years before that basic computer science), which has now led me to work with an amazing ex-Meta professor, collaborate with Google DeepMind researchers, and have interviews at amazing companies.
Having learned from all of my failures and successes, this video breaks down how I would learn machine learning all over again, focusing on the essentials and learning from the best resources.
Enjoy!

šŸ‘‰šŸ» Book a one-on-one call

šŸ‘‡šŸ» Links to resources šŸ‘‡šŸ»
==== Maths ====

==== ML/ DL ====

ā¬‡ļø Follow me on my other socials and feel free to DM questions! ā¬‡ļø
āš«āšŖ Medium:
šŸ¦Ā Twitter:

================== Timestamps ================
00:00 – Intro
00:40 – Python
01:29 – Maths
02:47 – ML Developer Stack
04:00 – Learn Machine Learning
06:06 – How To Really Get Good
=============================================

#ai #learning #machinelearning

source

47 Comments

  35. I am a bit new to Python, but list comprehension blew my mind – WTF? Who the hell designed this syntax? What sane person who wants to debug his code or come back to it a month later writes this? There are ways of writing really convoluted and complex unreadable code in other languages but that doesn't mean that you should use them. If you use this syntax then I feel sorry for your coworkers.

  39. Hello Boris sir, thank you for your youtube videos. Moreover, I want to ask whether buying M4max apple with 128 gb laptop for machine learning, AI and Data science or buy M4 max with 36 gb and use cloud for higher data computation. It would be our pleasure to have on best laptop idea for these categories.

  49. A ā€œMeta professorā€ is not a standard academic title. However, it might refer to one of the following:
    1. Meta as a Topic: A professor whose area of expertise involves ā€œmetaā€ topics, such as meta-cognition, meta-analysis, or meta-ethics, which explore the underlying frameworks or methodologies of a given field.
    2. Meta, the Company: A professor affiliated with or collaborating with Meta Platforms, Inc. (formerly Facebook), particularly in research fields like artificial intelligence, virtual reality, or social media studies.
    3. Self-Referential or Philosophical Context: Someone who studies or teaches about the nature of teaching, learning, or academic structures themselves (e.g., ā€œa professor about professorsā€).

  55. I enjoyed Andrew Ngā€™s ML specialization, but do with it was more hands on. Itā€™s great for learning about the algorithms but you donā€™t implement much of the projects yourself.

  59. Thats just basic stuff pandas is super basic used for converting dataset into readable form , performing data cleaning , exploratory data analysis . Learn scikit, tensor flow, learn algorithms for supervised n unsupervised learning like linear regression, logistic regression, DBSCAN , Isolation Forest, kMeans , Random Forests n many others

  63. hiii everyone , i have started learning ML about a month ago and i hear a lot of people saying that i need to be extremely good at SQL, but i am lost , like does anyone know topics should i cover and what projects can help understand the importance and usage of databases in ML, like i know that lots of times i am gonna have to download databses of the internet , but what else?

  79. To all the people watching this video thinking theyā€™re going to get a $300k salary self teaching: youā€™re not. Videos like this are misleading and send the wrong message to people.

  83. My journey which has been of ups and downs and I already know python
    1. Watching 3Blue1Brown videos on neural network. My first introduction to ML
    2. I tried to build a Multilayer perceptron from scratch for facial recognition and failed miserably, it took too long to compute as I was running it on my cpu or the logic is wrong somewhere. I'd to still figure that out (only using numpy)
    3. Used TF for facial recognition as it was required for my project
    4. Finally MNset handwritten digits were recognized from the scratch code and I used Cuda for optimization
    5. Learnt about CNN and all mathematics behind it
    6. Learning Pytorch now for lane detection using CNN

  85. Roadmap is super relevant, contemporary, practical and pragmatic. Coined expectations about schedule are just inadequate. Typical yesterday-neophyte's, "second project" misestimation. Basic python+basic numpy+basic pandas from scratch, in "a few weeks"? Seriously? Even in "a few months" it will be a challenging task for most of students.

Leave A Reply

Please enter your comment!
Please enter your name here