Related Posts
Coach
RKT, any thoughts?
TCS is the worst company one can be in.
Cheap mindset management cares more about timesheet,wfo,forceful mandatory trainings,imposing strict restrictions on laptop,access to personal docs only via vpn,no employee well being no salary hikes nth at all.90% projects filled with freshers coz they cost less.No WLB for good dedicated hard working employees as they are punished with more and more work, leaves rewards mostly for people who don’t work.But no issues kyunki job security hai 😂
Chief
Additional Posts in Data & Analytics Consultants
What is a data lake in basic terms?
Subject Expert
Got messaged by a C3 . ai recruiter. Read that wlb is bad and that the interview process is absurdly long, but the Glassdoor reviews are 4.2 and can't find actual hours worked posted by anyone. How's the culture really? I'd be aiming for DS consulting, something more functional but with DS/ML concepts as my differentiator.
C3.ai, Inc.
New to Fishbowl?
unlock all discussions on Fishbowl.
StatQuest on YouTube takes you through the basic workings of the algo. May need to watch some videos that are mentioned as prereqs. Plan about 2-3 hours if you want to take good notes.
As to more details, this 100 pager is as as ELI-in-biz-school-and-not-a-PhD as I've found. Written for a master's thesis, so it builds up all the statistical learning concepts rather than just throwing you in the deep end. Link in pieces cause Fishbowl hides comments with foreign domains...
https:// ntnuopen .ntnu .no/ntnu-xmlui/handle/11250/2433761
Not sure what to recommend for learning the API. The documentation is lacking, as per tradition. My favorite documentation search process with xgb (and sklearn for that matter) is when I have a question, someone already asked it on stack exchange, and the accepted answer is basically "idk maybe it does this" 🤦♂️ I guess my point is, good luck lol
This is super helpful, thanks.
Question for you: suppose you were drugged, blindfolded, and dropped in the middle of a desert. Your only way to survive is by finding high ground before time runs out. How would you approach finding it? (Assuming you can't take off the blindfold but can walk and the drugs have worn off.) Ofc the gradient (slope) in this case abstract to whatever optimizatiom problem you're trying to solve.
I can't remember the details enough especially to give a good eli5, but I hope this at least helps with understanding the objective. Please correct me as well if I'm wrong, its been a few years since I dabbled with any of it and I could be mixing it up with something else.
Been a while but let’s see if I remember correctly. Xgboost is more like bunch of stubby decision trees. Each subsequent tree seeks to minimize the remaining residuals.
Not gonna lie, I don’t understand your question at all. But XGBoost is literally its name, extreme gradient boosting. Don’t know if that helps, but it rips and works better than random forests most of the time.