Skip to content

Instantly share code, notes, and snippets.

View chyld's full-sized avatar
:electron:
import numpy

Chyld Medford chyld

:electron:
import numpy
View GitHub Profile
Data Science Workflow
- Collect Data
- Load into Pandas
- EDA
- Eliminate null values
- Filter - remove rows, cols
- Impute - replace missing values with mean, median, mode, 0, 1
- Advanced Impute - if missing values are real, use regression with
# COLORS
c0="\[\e[38;2;0;255;255m\]"
c1="\[\e[38;2;220;20;60m\]"
c2="\[\e[38;2;255;69;0m\]"
c3="\[\e[38;2;255;215;0m\]"
c4="\[\e[38;2;173;255;47m\]"
c5="\[\e[38;2;46;139;87m\]"
c6="\[\e[38;2;0;128;128m\]"
c7="\[\e[38;2;0;191;255m\]"
c8="\[\e[38;2;138;43;226m\]"
Host mint
HostName 172.16.237.200
User chyld
Host mintr
HostName 172.16.237.200
User chyld
LocalForward 127.0.0.1:3000 127.0.0.1:3000
LocalForward 127.0.0.1:3001 127.0.0.1:3001
LocalForward 127.0.0.1:3002 127.0.0.1:3002
LocalForward 127.0.0.1:3003 127.0.0.1:3003
name: py38
dependencies:
- python=3.8
- pip
- pip:
- ipython
- jupyterlab
- notebook
- voila
- numpy
# updated for ubuntu 19.10
sudo add-apt-repository ppa:git-core/ppa
sudo add-apt-repository ppa:jonathonf/vim
sudo add-apt-repository ppa:lazygit-team/release
sudo apt-get update
sudo apt-get install pydf ack inotify-tools curl htop tree build-essential neofetch tmux sqlite3 zsh silversearcher-ag hwinfo ranger screen tilix gparted jq openssh-server git vim lua5.3 fzy xclip dstat xsel lazygit colortest
sudo apt install snapd
sudo snap install bitwise
can a nn fit to non linear data
when scaling ... minmax or standard
what is forward prop
what is backward prop
what is a loss
draw a diagram, show a few weights... ask about partial derivitives
what is a batch size
what is epoch
when to stop training
what is an optimizer
You want to predict whether your 2011 Toyota Camry will get better than 35 mpg.
You collect data about various cars, and run it through a Logistic Regression model.
Here are the coefficients 0.00000035, 0.0002, 0.31 related to the features weight, hp, acceleration respectively.
When all features are zero, y is zero.
Your car, the Camry, weighs 2100 lbs, has 125 hp and has 9.1 acceleration.
Let's say your threshold for positive class is 0.95.
# ~/.bashrc: executed by bash(1) for non-login shells.
# see /usr/share/doc/bash/examples/startup-files (in the package bash-doc)
# for examples
# If not running interactively, don't do anything
case $- in
*i*) ;;
*) return;;
esac

Docker

  • docker run --rm -it ubuntu:18.10 /bin/bash
  • docker build -t chyld/ds:1 .
  • docker images
  • docker image rm <ID>
  • docker run --rm -it chyld/ds:1 /bin/bash --login

General

Pop Quiz:
You have a black box that takes in a sequence of numbers as follows:
input = 3.1, 54.2, 18.1, -44.9
... and it outputs these numbers.
output = 37.58, 129.56, 64.58, -48.82
If you were to input the number 310.5 into this black box:
- What would the output be?