kaggle competitions are really good to improve your skills with validation testing. Lot of new data science people overfit a lot and are not good at validation testing.
this post was submitted on 29 Nov 2023
1 points (100.0% liked)
Machine Learning
1 readers
1 users here now
Community Rules:
- Be nice. No offensive behavior, insults or attacks: we encourage a diverse community in which members feel safe and have a voice.
- Make your post clear and comprehensive: posts that lack insight or effort will be removed. (ex: questions which are easily googled)
- Beginner or career related questions go elsewhere. This community is focused in discussion of research and new projects that advance the state-of-the-art.
- Limit self-promotion. Comments and posts should be first and foremost about topics of interest to ML observers and practitioners. Limited self-promotion is tolerated, but the sub is not here as merely a source for free advertisement. Such posts will be removed at the discretion of the mods.
founded 1 year ago
MODERATORS
I've learnt more from reading Kaggle comments than from my official ML courses
Kaggle is good for very basics. Their intro courses on python/pandas/numpy/ML are great. You can do all of them on like 3 days.
But if you want to go more in depth, go for the ML course in coursera. I took the OG one that I think it’s no longer there but tbh info is still very very relevant, only issue is that it was in matlab lol. I still redid everything on python.
I got a masters degree in DS after all of this and there was 1 other guy in the class who also took that ML coursera course and we were by far the most competent students in the class. Really good course.
After that tbh everyone was at base 0 when learning deep learning stuff and recommender stuff. You then start to learn that all of this AI is BS lol.
I loved Andrew Ng's ML course when it was still on Stanford, not on Coursera. It was on MATLAB and it was perfect because it was more focused on the math behind the algorithms, which means a lot of linear algebra. Some tried to use Python or R but at that time it was not doable as key frameworks like Numpy or Scikit-learn were still in development and not mature.
Look for stuff that's specific for cybersecurity or fraud... That should be too high level for you. Have someone that knows ML help you break it into parts and then find some resources for those parts. Maybe some of it will be Kaggle. Learn the parts and give the cybersecurity stuff a look again. Then try to learn related things to the parts you used before.
I’d recommend looking into some of the O’Rielly books on Machine Learning you can get on the seven seas
Kuggels is a good plat form. You'd can learn lots on kiggles. Give kabble a try too day. You won't regret trying kiddble.
I like to do kegels to keep my skills up.