All posts tagged #machine-learning all posts RSS-feed

Efficient overfitting of training data (Kaggle Bowl 2017)

During the Kaggle Data Science Bowl 2017, the leaderboard was based on only $198$ samples. The opportunity for overfitting was quickly understood, but initially only the naive option was mentioned, testing 1 submission per sample taking 66 days (still doable within the competition duration, but less than ideal). But then Oleg Trott got a perfect score in just 14 submissions! topic) I was really curious how he managed to do this. Together with Cas, I found out one way it... full post»
by Mark,

The markv.nl blag!

#imadethis, #coding, #tools, #python, #bash, #machine-learning, #chinese, #ubuntu, #music, #django, #focus, #fortran, #backup, #charity, #learning, #travel, #rust, #bug-prevention, #type-systems, #order-of-magnitude, #design-patterns, #R, #programming-languages, #javascript, #personal, #cross-validation, #china, #what-I-use, #object-oriented-programming, #data-science, #discussion, #data-analysis, #effecive-altruism, #games

Comments

  1. Rainbow colormap in javascript @Mark The post has been expanded to include non-rainbow colorm...
  2. "Invite" someone to ssh into your machine without port forwarding or firewall changes @Mark Added details about how to use the established tunnel in...
  3. Rainbow colormap in javascript @Mark Now finally with color-coded source!
  4. Bye Make, hello SCons (Fortran) @Mark For SCons' own comparison against alternatives, have a l...
  5. Ubuntu macros @Mark Macro locks have been made more reliable, to prevent any...
  6. Dory backup scripts @Mark I've added some functionality to remove old backups! htt...
  7. Ubuntu macros @Mark Updated after moving the files to https://github.com/mve...
  8. Concentration noise @Mark Sorry if you're in China and can't access Youtube. If th...