ml-learning-doc

This document would contain whatever advice/resources I’ve found useful in my ML journey.

note that these are a lot of resources, which might feel overwhelming, even I haven't covered all of these things, so its fine, also you don't need these many resouces to learn, just search on youtube whatever topic you want to learn or are not able to understand, watch the video on that topic, ask whatever doubts you've from chatgpt/sonnet. yes, use ai to learn about ai (recursion hehe). also learn and implement stuff - don't just keep learning, because implementation should also be done (i did this mistake in the starting, just learned and didn't applied much - but i realised this: how one can be sure that he knows things if he has just consumed stuff and not created anything, right?, not saying that create another sota model, but just start implementing whatever you can).

Some really good advice:

• https://x.com/yoobinray/status/1844460463670886902

• https://x.com/daddynohara/status/1864513697424789574

• https://blog.gregbrockman.com/how-i-became-a-machine-learning-practitioner

• https://x.com/basvanopheusden/status/1955520236670816272

• Try out competitions of Kaggle (one at a time), learn and collaborate with other people :).

• Learn git from here: https://www.youtube.com/watch?v=mJ-qvsxPHpY

books

• download the python version: https://www.statlearning.com/ (to learn ml algos and math behind them)

• https://d2l.ai/ (to learn DL)

• this book by nitin mukesh for data science prep (interview prep guide).

• class notes of statistics and probability by professor amit kumar (note: there are some errors in this, so please be careful if you're reading or learning from this).

• books of packt, manning publications are also great (like sebestian's llm from scratch is pure goldmine, or modern computer vision with pytorch by packt publications).

• This document was quite helpful if you don’t prefer learning by books but still want to learn in structured manner: Mech - ML community resource list.pdf, ML RESOUCES-20250624T145356Z-1-001.zip (credits: Mitanshu)

• You can also refer from my notes (if you're able to understand my handwriting :)).

• Start writing about whatever you learn on websites like medium.com, because when you'll write down stuff, you'll realize that there are some gaps in your learning, and while pouring down your thoughts/understanding of concepts somewhere you'll try to fill those - and also other people might learn something useful from your writings, please note that don't write just for the sake of writing - or copy paste from chatgpt, write your original/raw thoughts!

Interesting and useful read:

• https://github.com/ayush-agarwal-0502/Placement-guidance-juniors/tree/main

• https://github.com/ayush-agarwal-0502/Placement-guidance-juniors/blob/main/ML%20Data%20Science%20syllabus.md

• zomato ds interview experience: https://freedium.cfd/https://medium.com/@AaryanAhuja11/zomato-on-site-job-interview-for-data-scientist-f1d50adef05f

dsa

• DSA: Also practice regularly some questions (in python) related to arrays, string, greedy, dp from leetcode.com and also SQL. Learn about DSA from these 2 guys:

• striver

• codestorywith mik

• cses problemest

Very good place to search about project ideas or learn:

• Hugging Face – Blog

• Fast.ai

• contributing a model to hugging face

• implement papers

• https://www.aleksagordic.com/blog

• https://magazine.sebastianraschka.com/

websites to refer to know about latest research papers/trends (ignore if you’re just starting in the ML domain, they might be overwhelming):

• papers with code

• hugging face daily papers

youtube channels/playlists

channels to learn theoritical stuff:

• https://www.youtube.com/watch?v=ZftI2fEz0Fw&list=PLKnIA16_Rmvbr7zKYQuBfsVkjoLcJgxHH&pp=0gcJCWMEOCosWNin (100 days of ml - this covers most important ml algos in depth)

• https://www.youtube.com/playlist?list=PLKnIA16_RmvYuZauWaPlRTC54KxSNLtNn (100 days of DL)

• https://www.youtube.com/@ArtemKirsanov (for maths)

• https://www.youtube.com/@statquest (for maths and ml algos)

• https://www.youtube.com/@fodoai (for maths and ml algos)

• https://www.youtube.com/@3blue1brown (for maths)

• https://www.youtube.com/@Eigensteve (for maths)

• https://www.youtube.com/playlist?list=PL2SOU6wwxB0uwwH80KTQ6ht66KWxbzTIo (for probability and statistics)

• https://www.youtube.com/playlist?list=PL49CF3715CB9EF31D (for linear algebra)

• https://www.youtube.com/@YannicKilcher (ml papers explanation)

• https://www.youtube.com/playlist?list=PLoROMvodv4rMiGQp3WXShtMGgzqpfVfbU (ML deep dive course by stanford, if you've already done 100 days of ml from campus x, that is enough, but if you want to learn more you can check out this playlist)

channels to learn practical stuff/implementation:

• sentdex - pythonprogramming.net (python and general ML)

• https://www.youtube.com/@coreyms (best place to learn stuff related to python)

• https://www.youtube.com/@NielsRogge

• https://www.youtube.com/@umarjamilai

• https://www.youtube.com/@AndrejKarpathy

• https://www.youtube.com/@geohotarchive (if you like programming/computer - you'll enjoy watching this channel)

• https://www.youtube.com/@FerdinandMom (for learning distributed training)

• https://www.youtube.com/watch?v=X2b_BD_LgwU

• https://www.youtube.com/watch?v=V8r__fXx7tU

• https://www.youtube.com/@Explaining-AI (very nice channel, covers stuff realted to CV indepth)