10:33 UTC

< July 2021 > Su Mo Tu We Th Fr Sa 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

- Console
- #amaranth-lang
- #armbian
- #armbian-allwinner
- #armbian-amlogic
- #armbian-broadcom
- #armbian-rockchip
- #armlinux
- #beagle
- #buildbot
- #commonlisp
- #crux
- #crux-arm
- #crux-devel
- #crux-social
- #crystal-lang
- #discferret
- #evennia
- #fedilinks
- #fedora-coreos
- #fedora-riscv
- #ffmpeg
- #ffmpeg-devel
- #foot
- #glasgow
- #hpy
- #hts
- #jruby
- #kisslinux
- #libreelec
- #linux-amlogic
- #linux-exynos
- #linux-mediatek
- #linux-rockchip
- #linux-ti
- #litex
- #maemo-leste
- #mailx
- #mew
- #mlpack
- #moin
- #nmigen
- #numbat
- #ocaml
- ##openfpga
- #openFPGALoader
- #openocd
- #openscad
- #openvswitch
- #osdev
- #picolisp
- #prjbureau
- #pypy
- #racket
- #radxa
- ##raspberrypi-internals
- #riscv
- #river
- #ruby
- #rust-embedded
- #sandstorm
- #scopehal
- #solvespace
- #Speedsaver
- ##stm32-rs
- #tegra
- #titandev
- #u-boot
- ##yamahasynths
- #yocto
- #yosys
- #zeppe-lin

rcurtin_irc changed the topic of #mlpack to: mlpack: a scalable machine learning library (https://www.mlpack.org/) -- channel logs: https://libera.irclog.whitequark.org/mlpack -- NOTE: messages sent here might not be seen by bridged users on matrix, gitter, or slack

< RishabhGarg108Ri>
Hey @ryan:ratml.org, I wanted to know why we are taking square root in this line https://github.com/mlpack/mlpack/blob/c8647b2bfef68dfd29a5a8612813f92cf2b79de0/src/mlpack/methods/decision_tree/multiple_random_dimension_select.hpp#L44?

<RishabhGarg108Ri>
Hey @ryan:ratml.org, I wanted to know why we are taking square root in this line https://github.com/mlpack/mlpack/blob/c8647b2bfef68dfd29a5a8612813f92cf2b79de0/src/mlpack/methods/decision_tree/multiple_random_dimension_select.hpp#L44?

< RishabhGarg108Ri>
What is the intuition behind it? Why not `numDimensions = dimensions` or `numDimensions = dimensions / 2`? Is it square root a heuristic?

<RishabhGarg108Ri>
What is the intuition behind it? Why not `numDimensions = dimensions` or `numDimensions = dimensions / 2`? Is it square root a heuristic?

< RishabhGarg108Ri>
* What is the intuition behind it? Why not `numDimensions = dimensions` or `numDimensions = dimensions / 2`? Is taking square root a heuristic?

<RishabhGarg108Ri>
* What is the intuition behind it? Why not `numDimensions = dimensions` or `numDimensions = dimensions / 2`? Is taking square root a heuristic?

< rcurtin[m]>
RishabhGarg108 (RishabhGarg108): sorry for the slow response. I went digging for the source this morning, but I think the original Random Forests paper by Leo Breiman suggested using sqrt(d) here

<rcurtin[m]>
RishabhGarg108 (RishabhGarg108): sorry for the slow response. I went digging for the source this morning, but I think the original Random Forests paper by Leo Breiman suggested using sqrt(d) here

< rcurtin[m]>
Wikipedia suggests it's page 592 in "The Elements of Statistical Learning": https://en.wikipedia.org/wiki/Random_forest (see "From bagging to random forests")

<rcurtin[m]>
Wikipedia suggests it's page 592 in "The Elements of Statistical Learning": https://en.wikipedia.org/wiki/Random_forest (see "From bagging to random forests")

< RishabhGarg108Ri>
I checked the book. It says `floor(sqrt(p))` for classification and `floor(p / 3)` for regression.