---

Do you want to get updates? Please join Data Sci­ence & Machine Learn­ing Newslet­ter Linked Group

---

• Visu­al­iz­ing Incom­plete and Miss­ing Data
  • “… But a lot of the time (most of the time?), the data you work with is not com­plete. There is miss­ing data. … What do you do when this hap­pens? … Here are some solu­tions to get you headed in the right direction.”
• A Gen­tle Intro­duc­tion to Vec­tors for Machine Learn­ing
  • “Vec­tors are a foun­da­tional ele­ment of lin­ear alge­bra. Vec­tors are used through­out the field of machine learn­ing in the descrip­tion of algo­rithms and processes such as the tar­get vari­able (y) when train­ing an algo­rithm. In this tuto­r­ial, you will dis­cover lin­ear alge­bra vec­tors for machine learning.”
• Top 4 open source alter­na­tives to Google Ana­lyt­ics
  • “Many busi­nesses of all sizes use Google Ana­lyt­ics. But if you want to keep con­trol of your data, you need a tool that you can con­trol. You won’t get that from Google Ana­lyt­ics. Luck­ily, Google Ana­lyt­ics isn’t the only game on the web.”
• Info­graphic: Poker & AI: The Rise Of Machines Against Humans
• Autoscal­ing deep learn­ing clus­ters on AWS with Kuber­netes and RiseML
  • “A lit­tle known fea­ture of the RiseML installer is it’s abil­ity to cre­ate deep learn­ing clus­ters with autoscal­ing sup­port on AWS. Autoscal­ing can help sig­nif­i­cantly reduce GPU bills and increase com­pute capac­ity dur­ing peak times by auto­mat­i­cally launch­ing and ter­mi­nat­ing EC2 instances based on demand.”
• Cat­Boost is an open-source gra­di­ent boost­ing library with cat­e­gor­i­cal fea­tures sup­port
  • “Cat­Boost ver­sion 0.6 has a lot speedups and improve­ments. Most valu­able improve­ment at the moment is the release of indus­try fastest infer­ence implementation.”
• The Matrix Cal­cu­lus You Need For Deep Learn­ing
  • “This paper is an attempt to explain all the matrix cal­cu­lus you need in order to under­stand the train­ing of deep neural net­works. We assume no math knowl­edge beyond what you learned in cal­cu­lus 1, and pro­vide links to help you refresh the nec­es­sary math where needed.”
• https://github.com/eriklindernoren/NapkinML
  • A tiny lib with pocket-sized imple­men­ta­tions of machine learn­ing mod­els in NumPy.
• https://www.gnu.org/software/datamash/
  • GNU data­mash is a command-line pro­gram which per­forms basic numeric,textual and sta­tis­ti­cal oper­a­tions on input tex­tual data files.